Recognize the exact 1963 AMTCL annual meeting volume

Merge pull request '限定ACL1978与TINLAP卷的来源关系及非论文条目' (#42 ) from codex/acl-early into main
Reviewed-on: #42
2026-11-09 23:21:53 +08:00 · 2026-11-09 13:42:18 +00:00 · 2026-11-09 13:42:00 +00:00 · 2026-11-09 20:50:29 +08:00 · 2026-11-09 17:20:27 +08:00 · 2026-11-09 09:02:49 +00:00
22 changed files with 15940 additions and 9 deletions
@@ -50,6 +50,9 @@ jobs:
              subprocess.run([python,'-I','-B','-m','paper_atlas','collect','cvf','--help'],cwd=outside,check=True)
              subprocess.run([python,'-I','-B','-m','paper_atlas','collect','www-archive','--help'],cwd=outside,check=True)
              subprocess.run([python,'-I','-B','-m','paper_atlas','collect','aamas-proceedings','--help'],cwd=outside,check=True)
+              subprocess.run([python,'-I','-B','-m','paper_atlas','collect','aamas-proceedings','--year','2025','--snapshots',str(src/'tests/fixtures/aamas2025'),'--out',str(root/'installed-aamas2025')],cwd=outside,check=True)
+              subprocess.run([python,'-I','-B','-m','paper_atlas','collect','iclr-archive','--help'],cwd=outside,check=True)
+              subprocess.run([python,'-I','-B','-m','paper_atlas','collect','acl','--help'],cwd=outside,check=True)
              subprocess.run([python,'-I','-B','-m','paper_atlas','maintain','postgres','--help'],cwd=outside,check=True)
              subprocess.run([python,'-I','-B','-m','paper_atlas','maintain','edition-tracks','--help'],cwd=outside,check=True)
              subprocess.run([python,'-I','-B','-m','paper_atlas','maintain','joint-editions','--help'],cwd=outside,check=True)
@@ -6,8 +6,12 @@ Paper Atlas 将 arXiv、AI 会议论文集及公开投稿决定整理到可追

 项目仍在建设中。候选范围包含 43 个会议（含 ACM MM、MLSys），尚未完成全部历史数据和最终去重复核。实际进度以数据库、来源清单和覆盖报告为准，而不是仓库文件数量。

+历史 ICLR 2014–2016 官方录用页由 `collect iclr-archive` 适配，保留主会/workshop及重复展示证据；运行和身份核验边界见 [历史档案说明](docs/coverage/iclr-archive-2014-2016.md)。
+
 ## 项目结构

+ACL 1963 的 AMTCL 年会摘要与 1978 的 TINLAP-2 来源关系均按精确卷适配，保留原系列、出版形式和不明逐篇角色；不将 EarlyMT 全系列映射到 ACL，1965 的替代关系也不转成全局别名。来源、排除条目和恢复边界见 [ACL 早期来源核查](docs/coverage/acl-early-sources-2026-11-09.md)。
+
 ```text
 paper_atlas/
  cli.py                 统一命令入口
@@ -64,7 +68,7 @@ paper-atlas collect colm --cache /data/paper-index/raw/http-cache \

 EWSL、EPS、ECP、EWSP 的[具体前身卷与完整快照要求](docs/coverage/historical-predecessors-2026-11-09.md)单独核验，保留原始系列及事件/出版年份；旧候选文件缺少 ECP/EWSP 时会明确提示，避免静默遗漏。

-AAMAS 的[官方论文集采集器](docs/aamas-proceedings.md)目前核验了 2012 版式，可从缓存或已校验快照采集标题、作者及页面提供的摘要；发表形式单独保存，全文/扩展摘要中未明确的投稿 track 保持待明确，不根据会话主题推定主会轨道。
+AAMAS 的[官方论文集采集器](docs/aamas-proceedings.md)支持已核验的 2012、2025 版式，可从缓存或已校验快照采集元数据。2025 年目录含 476 条非 keynote 出版物，保留完整作者和机构，HTML 未提供摘要时留空；发表形式与投稿 track 分开，不根据主题推定主会轨道或研究类型。

 `paper-atlas collect --help` 查看来源；`paper-atlas import --help` 查看导入方式。完整批次步骤见[使用指南](docs/usage.md)。

@@ -1,6 +1,14 @@
 # AAMAS 官方论文集元数据

-`paper_atlas.collectors.aamas` 是按需运行的采集适配器，通过统一入口 `collect aamas-proceedings` 使用。目前只支持已独立核验的 **AAMAS 2012** HTML 版式；其他年份需先核对官方目录和逐篇结构，再增加适配，不能仅替换年份。
+`paper_atlas.collectors.aamas` 是按需运行的采集适配器，通过统一入口 `collect aamas-proceedings` 使用。支持已独立核验的 **AAMAS 2012、2025** HTML 版式；其他年份需先核对官方目录和逐篇结构，再增加适配，不能仅替换年份。2025 解析单独封装在 `paper_atlas.collectors.aamas2025`。
+
+2025 版式使用[官方首页](https://www.ifaamas.org/Proceedings/aamas2025/)和[出版目录](https://www.ifaamas.org/Proceedings/aamas2025/forms/contents.htm)。原始目录共 479 条：3 keynote、260 research paper、148 extended abstract、12 Blue Sky、25 doctoral、22 Demo 栏目、9 JAAMAS。排除 keynote，输出 476 条；Demo 栏目中的 p3000 是明确标题的 COMP24 竞赛报告，发表年份仍为 2025，格式记 `competition_report`、投稿 track 留作 `proceedings_unspecified`，保留原 Demo 栏目证据。其余输出按 track 分组为 409 unspecified、12 blue_sky、25 doctoral、21 demo、9 jaamas。
+
+作者按逐篇 BR 行解析，斜体机构原文另存，保留 Unicode、顺序及完整名单。HTML 没有提供摘要或逐篇 DOI，故 `abstract=null`、标识字典为空；只保存 PDF 链接，不下载研究 PDF。不能把 Research Paper Track 的所有条目推成 main，不能把目录缺席推为 rejected/withdrawn，也不能把出版格式标签当研究类型。
+
+2025 用法：`paper-atlas collect aamas-proceedings --year 2025 --snapshots /data/review/aamas2025 --out /data/paper-index/collected/aamas-2025`；在线模式改用 `--cache /data/paper-index/cache/aamas2025`，只请求上述两个 HTML 页面。快照目录包含原始 `index.html`、`contents.html` 和 `receipts.json`，凭据必须列出各原始 URL、文件、字节数、SHA256、`http_status=200` 和带时区 `at`。已恢复的凭据需保留 `at_precision=filesystem_write_time_recovered` 等说明，不能伪造 HTTP 头或请求时间。2026-11-09 保存的两个页面通过哈希校验；独立录用清单 URL 返回 403，失败保留，不绕过或依赖其完成采集。
+
+2025 命令同样使用目录锁、临时目录、最后写入采集报告。五个 JSONL/manifest 均保持 `partial`，记录摘要缺失数量；它不连接数据库。首次采集需独立核验全部标题/作者/栏目和 source IDs，随后做 PostgreSQL 身份预检、必要的逐条版本审核、原生导入与独立验收。若失败且没有完成报告，应先检查原始响应和错误，再使用新输出目录重跑，不重放已提交批次。代码验证、原始目录盘点和采集成功均不代表已导入生产或最终去重完成。

 ## 来源与范围

@@ -0,0 +1,64 @@
+# ACL 早期年份：会议存在、出版目录与逐篇资料分开核验
+
+固定普查中 ACL 1963–1978 仍有年份缺口。不能因为会议年表连续，就假设每年都有独立的正式论文集；也不能因为当前目录始于 1979，就断言此前没有摘要、通知或其他公开资料。本次不修改固定的 1,266 个目标年份基准。
+
+## 1978 年：限定 T78-1 的来源关系
+
+[同期年会通知](https://aclanthology.org/J77-4001.pdf)的 PDF 第 9–10 页（印刷页码 9–10）明确说明，1978 年 ACL 年会采用第二届 Theoretical Issues in Natural Language Processing 的形式，地点为 Urbana/Champaign，日期为 7 月 25–27 日。通知中的 ACM 联合赞助确认仍处于待确认状态，不能将该预告改写为已完成的行政决定。
+
+[1982 年 Donald E. Walker 的历史回顾](https://aclanthology.org/www.mt-archive.info/70/ACL-1982-Walker-1.pdf)第 3 页（印刷页码 91）的年会表及缩写说明，独立对应了 1978 年的时间、地点与 TNLP/TINLAP-2 关系。两份扫描件相关页已渲染并逐页目视核对，不仅依靠 OCR。
+
+[官方卷目录](https://aclanthology.org/volumes/T78-1/)与固定 XML `T78.xml` 保留原系列名 TINLAP、完整卷名、1978 年及编辑 David L. Waltz。适配仅接受精确的 collection `T78`、volume `1` 和上述字段组合；关键元数据变化会拒绝处理，等待重新核查。
+
+- 规范事件名使用 `ACL+TINLAP`，原系列名 `TINLAP`、原始 XML 的 `venue_codes: [tinlap]` 及证据均保留。
+- ACL/TINLAP 别名只挂在这一规范事件系列上，不把所有 TINLAP 年份都算成 ACL。T75、T87 及 T78 其他卷没有此例外。
+- 通知描述了邀请与提交论文两种参与方式，XML 没有逐篇角色。统一暂存为 `proceedings_unspecified`，不猜测每篇属于 invited、main 或 workshop。
+- XML 共 36 条条目，其中 T78-1036 是无作者的场次讨论问题列表。精确识别并在 manifest 的 `excluded_materials` 保留其标识、标题与原因；其余 35 条作为具名贡献整理。该排除不推广到所有无作者条目，已识别条目发生变化也须重新核查。
+- 全部 35 条 XML 均无摘要。出版目录中的 accepted 是来源观察；通知提及录用／不录用通知时间，不提供任何可识别的拒稿决定，不能制造 rejected 或 withdrawn。
+
+采集器仅生成元数据文件，不写业务数据库。真实来源逐条验收、PostgreSQL 身份预检、串行导入与独立验收仍是后续门槛。35 条来源不等于 35 个新增论文实体，更不等于完成全库去重。
+
+## 1963 年：AMTCL 旧名称与 EarlyMT 分类分开保留
+
+[ACL 官方历史介绍](https://2023.aclweb.org/participants/about_acl/)确认 AMTCL 是协会旧名称，1968 年改称 ACL。[1963 官方卷目录](https://aclanthology.org/volumes/1963.earlymt-1/)和固定 XML `1963.earlymt.xml` 明确记录 AMTCL 年会、Denver、1963 年 8 月 25–26 日。源目录将该卷归在 EarlyMT；这是资料分类，不能把其他 EarlyMT 会议全部变成 ACL。
+
+适配仅接受 collection `1963.earlymt`、volume `1`，并核对旧名称全称、年份、地点、日期和原始 venue code。规范会议为 ACL，保留 AMTCL 别名及原始 `earlymt` 分类。关键字段变化或重复时拒绝处理，等待重新审核。
+
+34 条均有作者和摘要，来源是出版摘要清单，记录 `published_abstract` 出版形式及 `acl_anthology:publication_format` 标签；逐篇 track 未提供，使用 `proceedings_unspecified`。`accepted` 只表示被列入该出版清单，不推断现代 main track、原始评审决定或任何拒稿记录。
+
+XML commit 与下文相同，SHA256 为 `4a4626d792ac507de0281cb358769f7246f4a0d6b6a9176ac02e61545c1635fc`。原始 fixture 为 `tests/fixtures/acl-1963-earlymt.xml`。官方历史 HTML SHA256 为 `0f6bc260c2b34c9e2a51d971aeb43c3270a5b15b6db1d6d41e50669f316345f3`，响应回执与固定 XML 保存在服务器 `/data/paper-index/collected/acl1963-scope-evidence-v1`。另有卷目录 HTML 回执保存在 `/data/paper-index/collected/historical-primary-candidates-20261109-v1`。未下载论文 PDF。
+
+使用下文相同采集命令，将 `--collection` 改为 `1963.earlymt`，输出到新的独立目录。采集结果需逐条验收和 PostgreSQL 身份预检，再进入增量导入；34 条来源不等于 34 个新增论文实体，代码测试通过也不代表生产覆盖已补齐。
+
+## 1965 年：替代关系不改写为别名
+
+Walker 回顾第 1 页（印刷页码 89）说明，1965 年 ICCL（后来称 COLING）替代了当年的年会。[官方 C65-1 目录](https://aclanthology.org/volumes/C65-1/)及 XML 仍以 COLING 标记该卷。这是限定年份的历史关系，不证明 COLING 与 ACL 是两个可全局互换的名字。
+
+本次保留 C65 原有 `COLING` 归属，不添加 ACL 全局别名、不重放已导入记录、不将某一年的日历关系推广到整个系列，也不据此悄悄减少固定普查缺口。其他 1963–1977 年资料仍须逐年核查；本次不能证明这些年份整体属于非归档例外。
+
+## 固定来源与恢复
+
+服务器证据目录：`/data/paper-index/collected/acl-early-source-evidence-v1`。五份官方响应均 HTTP 200，另有两份来自既有官方 XML checkout 的文件，已与 Git 对象逐字节核对。未下载论文全文集合；两份 PDF 是本次历史／行政取证资料，不进入论文全文库。
+
+| 文件 | SHA256 |
+|---|---|
+| `acl-venue.html` | `c60f64cb4d72e677aeb84a5844cde28a8d50579e1995f08e8c7801eed1e3fac5` |
+| `acl-history-1982.pdf` | `35c21bd4983975bcaccbb94bd8ab91c9019fc190feec1791255407cf18a7e49a` |
+| `acl-1978-notice.pdf` | `14b5deb471c55afb07d789529b95d0462a2a1a77df3dfba7fa1f5d3ab75eb411` |
+| `tinlap1978-volume.html` | `7e20d709f72f415819e4d17e3674b19dc7d871eabcf0cf8b4935f0392d775729` |
+| `coling1965-volume.html` | `0d2f2f49f2a311f9c0b80a8832f6476eb32f94f35abf36f6d2f2e10063ac880f` |
+| `T78.xml` | `f712f94cff0cba05f786b7ac31dad2ea9408fde2a17a04531130df623f57dd84` |
+| `C65.xml` | `572791610e20b7c6045d984d968e73a18a144bcb5c3da64112001c781826808e` |
+
+XML commit：`a7be623d01c48c61405dec31c871121cdbed5a01`。测试 fixture `tests/fixtures/acl-t78.xml` 是该原始文件，不是手工模拟论文；不包含 PDF。PDF 哈希建立完整性，不单独证明语义，语义依据是上述具体页面。
+
+在服务器使用既有命令，固定 checkout 并输出到新目录：
+
+```sh
+paper-atlas collect acl \
+  --checkout /data/paper-index/raw/metadata-checkouts/acl-anthology \
+  --collection T78 \
+  --out /data/paper-index/collected/acl-t78-source-v1
+```
+
+该命令会校验／更新官方 metadata checkout，实际 commit 必须以生成清单为准；严格重现本次取证时，使用已冻结的 XML 和上述 commit，不把更新后的内容冒充旧快照。不要将无 collection 限制的全部历史输出直接作为增量队列。采集完成不代表导入完成，恢复前先检查当前单元和报告，不因日志安静重启已执行的写入。
@@ -0,0 +1,44 @@
+# ICLR 2014–2016 官方历史录用档案
+
+此适配器是长期维护工具的一部分，按需运行。只识别已经独立核对的三种页面布局；后续页面发生变化必须重新核查，不能放松计数或轨道边界以使采集成功。
+
+| 官方页面 | 页面链接次数 | 整理后的不同记录 |
+|---|---:|---:|
+| [2014 Conference Proceedings](https://iclr.cc/archive/2014/conference-proceedings/) | 35 | main 35 |
+| [2015 accepted-main](https://iclr.cc/archive/www/doku.php%3Fid=iclr2015:accepted-main.html) | 116 | main 31、workshop 74 |
+| [2016 accepted-main](https://iclr.cc/archive/www/doku.php%3Fid=iclr2016:accepted-main.html) | 80 | main 80 |
+
+2015 的 11 篇 oral 全部再次出现在 poster 段落中。相同年份、轨道及 arXiv 标识的重复展示只生成一条来源记录，但每次展示的标题、作者显示字符串、链接、版本和段落均保留在 `source_variants`。两处标题或解析作者冲突时停止采集、等待人工核查。不同年份或不同轨道不合并投稿。
+
+2014 的作者以分号列出，一条有末尾分号，一条标题和作者斜体块不在普通段落中，部分作者段落拆成多个格式片段；必须读取完整作者段落或独立斜体块，不能只读首个格式片段。2016 的代码链接不属于作者。保留原文，解析结果不替代来源证据。oral、poster 作为 `iclr_archive:presentation` 标签，不改成独立投稿轨道。arXiv 版本留在来源变体中，稳定标识使用去版本号的编号。
+
+## 来源快照与运行
+
+2026-11-09 已冻结的官方响应 SHA256：
+
+- 2014：`a33858176fbffae0066ed3c10310f7dd98a66d12a36334b6339ef1f915ef07af`
+- 2015：`131491663742cb084662a4804009b8318e0fa7bc184d4a9044d16dd28f05b888`
+- 2016：`7320b8652431793b6ea0743f871ee446bf08b6361ccd4e421b6b53449e71f4f6`
+
+在服务器运行，输出新目录，不下载论文 PDF：
+
+```sh
+paper-atlas collect iclr-archive --year 2015 \
+  --snapshots /data/paper-index/collected/historical-next-source-evidence-v1 \
+  --out /data/paper-index/collected/iclr-archive-source-v1/2015
+
+# 或重新下载官方 HTML，并保留 HTTP 缓存与收据
+paper-atlas collect iclr-archive --year 2016 \
+  --cache /data/paper-index/cache/iclr-archive \
+  --out /data/paper-index/collected/iclr-archive-source-v1/2016
+```
+
+冻结输入目录必须包含 `receipts.json`，每项有精确 `url`、HTTP `status: 200`、相对 `file`、`bytes`、`sha256` 和带时区的 `finished_at`。读取时重新核对尺寸、哈希和路径；拒绝越界及指向目录外的软链接。已完成输出不可覆盖，应检查报告后另建明确版本目录。解析或计数失败时不生成完成报告，也不导入数据库。
+
+输出分轨道 JSONL、manifest、原始 HTML 证据和年度采集报告。仅支持这三年已审查的页面。2014 workshop、其他未列轨道、逐篇摘要及公开拒稿范围尚未通过这些页面核验，不能据此声称完整。缺席不代表 rejected、withdrawn 或从未被录用。
+
+## 导入前后的边界
+
+这三页合计 220 条不同来源记录，不等于新增 220 篇论文；同一工作可能已存在于 arXiv 或其他会议。先核验来源字段与原始页面，再执行 PostgreSQL 只读预检，比较稳定标识、标题、作者和原始来源。标题/作者差异、版本关联及冲突必须保留证据并明确评估；不能仅靠标题猜测身份，也不能覆盖原有摘要。
+
+采集器不连接数据库。正式导入使用既有 `import records`，必须有新的生产保护基线和独立验收；全部导入结束后仍需全库去重和历史覆盖审计。来自官方录用清单的 accepted 是清单观察，不是原始审稿决定签名或决定日期。
@@ -13,6 +13,7 @@ COMMANDS = {
    ('download',): ('paper_atlas.io.snapshots', {}, '下载并校验固定版本快照'),
    ('collect','pmlr'): ('paper_atlas.collectors.pmlr', {}, 'PMLR 官方元数据'),
    ('collect','acl'): ('paper_atlas.collectors.acl', {}, 'ACL Anthology XML'),
+    ('collect','iclr-archive'): ('paper_atlas.collectors.iclr_archive', {}, '核对ICLR 2014–2016官方历史录用档案'),
    ('collect','iclr-program'): ('paper_atlas.collectors.iclr', {}, 'ICLR 官方公开程序'),
    ('collect','iclr-snapshot'): ('paper_atlas.collectors.openreview', {}, 'ICLR 公开投稿与决定快照'),
    ('collect','public-openreview'): ('paper_atlas.collectors.public_openreview', {}, '核验其他会议的公开拒稿及撤稿标签'),
@@ -136,6 +136,9 @@ def read_pages(cache=None, snapshots=None):


 def _collect(year, out, *, cache=None, snapshots=None):
+    if year == 2025:
+        from paper_atlas.collectors.aamas2025 import collect as collect_2025
+        return collect_2025(out, cache=cache, snapshots=snapshots)
    if year != 2012:
        raise ValueError('Only the independently reviewed AAMAS 2012 layout is supported')
    pages, receipts = read_pages(cache, snapshots)
@@ -190,8 +193,8 @@ def _collect(year, out, *, cache=None, snapshots=None):


 def collect(year, out, *, cache=None, snapshots=None):
-    if year != 2012:
-        raise ValueError('Only the independently reviewed AAMAS 2012 layout is supported')
+    if year not in (2012, 2025):
+        raise ValueError('Only independently reviewed AAMAS 2012/2025 layouts are supported')
    out = Path(out)
    out.mkdir(parents=True, exist_ok=True)
    with (out/'aamas.collect.lock').open('a') as lock:
@@ -201,7 +204,7 @@ def collect(year, out, *, cache=None, snapshots=None):

 def main(argv=None, *, prog=None):
    parser = argparse.ArgumentParser(prog=prog)
-    parser.add_argument('--year', type=int, choices=[2012], required=True)
+    parser.add_argument('--year', type=int, choices=[2012, 2025], required=True)
    parser.add_argument('--out', required=True)
    source = parser.add_mutually_exclusive_group(required=True)
    source.add_argument('--cache')
@@ -0,0 +1,183 @@
+"""Reviewed AAMAS 2025 publisher TOC; HTML metadata only, never paper PDFs."""
+from collections import Counter, defaultdict
+import hashlib
+import json
+from pathlib import Path
+import re
+import tempfile
+
+from paper_atlas.db.submissions import timestamp
+from paper_atlas.io.html import Document, Node, clean
+from paper_atlas.io.http import fetch
+
+ROOT = 'https://www.ifaamas.org/Proceedings/aamas2025/'
+URLS = {'index': ROOT, 'contents': ROOT+'forms/contents.htm'}
+TITLE = 'AAMAS 2025: Proceedings of the 2025 International Conference on Autonomous Agents and Multiagent Systems'
+SECTIONS = {
+    '1': ('Keynote Talks', 'keynote', 'keynote', 3),
+    '2': ('Research Paper Track', 'full_paper', 'proceedings_unspecified', 260),
+    '3': ('Extended Abstracts', 'extended_abstract', 'proceedings_unspecified', 148),
+    '4': ('Blue Sky Ideas', 'blue_sky_paper', 'blue_sky', 12),
+    '5': ('Doctoral Consortium', 'doctoral_abstract', 'doctoral', 25),
+    '6': ('Demo Track', 'demo', 'demo', 22),
+    '7': ('JAAMAS Track', 'journal_presentation', 'jaamas', 9),
+}
+COMPETITION_TITLE = '[COMP24] The Automated Negotiating Agents Competition (ANAC) 2024 Challenges and Results'
+PAPER = re.compile(r'\.\./pdfs/p([0-9]+)\.pdf')
+
+
+def one(items, label):
+    if len(items) != 1:
+        raise ValueError('Expected one '+label)
+    return items[0]
+
+
+def parse_contents(html, evidence):
+    doc = Document(html)
+    if [clean(n.text()) for n in doc.root.walk() if n.tag == 'title'] != [TITLE]:
+        raise ValueError('Unexpected AAMAS proceedings title/year')
+    for paragraph in doc.root.walk():
+        if paragraph.tag != 'p' or any(n is not paragraph and n.tag == 'p' for n in paragraph.walk()):
+            continue
+        anchors = [n.attrs['name'] for n in paragraph.walk() if n.tag == 'a' and n.attrs.get('name') in SECTIONS]
+        if anchors and (len(anchors) != 1 or clean(paragraph.text()) != SECTIONS[anchors[0]][0]):
+            raise ValueError('Publisher section heading differs from reviewed layout')
+    section = None
+    headers = []
+    records, excluded, counts, seen = [], [], Counter(), set()
+    for node in doc.root.walk():
+        if node.tag == 'a' and node.attrs.get('name') in SECTIONS:
+            section = node.attrs['name']
+            headers.append(section)
+        if node.tag != 'p' or any(n is not node and n.tag == 'p' for n in node.walk()):
+            continue
+        links = [n for n in node.walk() if n.tag == 'a' and PAPER.fullmatch(n.attrs.get('href', ''))]
+        if not links:
+            continue
+        link = one(links, 'paper link per paragraph')
+        if section not in SECTIONS:
+            raise ValueError('Paper outside a reviewed publisher section')
+        page = int(PAPER.fullmatch(link.attrs['href'])[1])
+        if page in seen:
+            raise ValueError('Repeated publisher paper/page ID')
+        seen.add(page)
+        title = clean(link.text())
+        if not title:
+            raise ValueError('Empty paper title')
+        # Names are outside italicized affiliations and separated by explicit BRs.
+        font = one([n for n in node.walk() if n.tag == 'font' and n.attrs.get('color', '').lower() == '#222222'], 'ordered author block')
+        authors, affiliations = [], []
+        name, affiliation = '', ''
+        def flush():
+            nonlocal name, affiliation
+            if clean(name) or clean(affiliation):
+                if not clean(name):
+                    raise ValueError('Affiliation without author')
+                authors.append(clean(name)); affiliations.append(clean(affiliation))
+            name, affiliation = '', ''
+        for child in font.children:
+            if isinstance(child, str):
+                name += child
+            elif child.tag == 'br':
+                flush()
+            elif child.tag == 'i':
+                affiliation += child.text()
+            else:
+                raise ValueError('Unreviewed author markup: '+child.tag)
+        flush()
+        if not authors:
+            raise ValueError('Empty author list')
+        heading, publication_format, track, _ = SECTIONS[section]
+        counts[section] += 1
+        if section == '1':
+            excluded.append({'publisher_page': page, 'title': title, 'authors': authors, 'reason': 'publisher keynote talk, not a research-paper submission'})
+            continue
+        note = 'Explicit publisher section; section topics are not research types.'
+        if track == 'proceedings_unspecified':
+            note = 'Published format is known; individual main/special submission track is not provided.'
+        if page == 3000:
+            if section != '6' or title != COMPETITION_TITLE:
+                raise ValueError('Competition report differs from reviewed publisher entry')
+            publication_format, track = 'competition_report', 'proceedings_unspecified'
+            note = 'Publisher places this competition report in Demo Track; no individual demonstration submission is inferred.'
+        records.append({
+            'source': 'aamas_proceedings', 'external_id': '2025/p'+str(page),
+            'url': URLS['contents']+'#'+section, 'title': title, 'abstract': None,
+            'authors': authors, 'identifiers': {}, 'venue': 'AAMAS', 'venue_aliases': [],
+            'year': 2025, 'track': track, 'status': 'accepted',
+            'observed_at': evidence['observed_at'],
+            'raw_decision': 'listed in official AAMAS 2025 published proceedings '+heading,
+            'tags': [{'namespace': 'aamas_proceedings:publication_format', 'value': publication_format}],
+            'metadata': {'publisher_page': page, 'paper_url': ROOT+'pdfs/p'+str(page)+'.pdf',
+                         'section_url': URLS['contents'], 'section_anchor': section,
+                         'section_heading': heading, 'publication_format': publication_format,
+                         'author_affiliations': affiliations, 'source_evidence': evidence,
+                         'track_scope_note': note, 'decision_date_precision': 'not_supplied',
+                         'missing_abstract_reason': 'Publisher HTML supplies no abstract; research PDF not downloaded.',
+                         'publication_evidence': 'Published proceedings entry, not an original decision note.'}})
+    if headers != list(SECTIONS) or dict(counts) != {k: v[3] for k, v in SECTIONS.items()}:
+        raise ValueError('Publisher section inventory differs from reviewed 2025 layout')
+    if len(records) != 476 or len(excluded) != 3 or sum(r['metadata']['publication_format'] == 'competition_report' for r in records) != 1:
+        raise ValueError('Unexpected publication partition')
+    return records, excluded
+
+
+def read_pages(cache, snapshots):
+    if bool(cache) == bool(snapshots):
+        raise ValueError('Choose exactly one cache or frozen snapshot directory')
+    frozen = json.loads((Path(snapshots)/'receipts.json').read_text())['receipts'] if snapshots else []
+    pages = {}
+    for key, url in URLS.items():
+        if snapshots:
+            r = one([r for r in frozen if r['url'] == url], 'frozen response receipt')
+            filename = Path(r['file']); base = Path(snapshots).resolve()
+            if filename.is_absolute() or '..' in filename.parts or not (base/filename).resolve().is_relative_to(base) or r.get('http_status') != 200:
+                raise ValueError('Unsafe or unsuccessful publisher response')
+            data = (base/filename).read_bytes()
+            if len(data) > 8*1024**2 or len(data) != r['bytes'] or hashlib.sha256(data).hexdigest() != r['sha256']:
+                raise ValueError('Frozen publisher response checksum/size mismatch')
+            observed = r['at']
+        else:
+            data = fetch(url, cache, 8*1024**2, delay=1)
+            r = json.loads((Path(cache)/hashlib.sha256(url.encode()).hexdigest()).with_suffix('.json').read_text())
+            observed = r['retrieved_at']
+        timestamp(observed)
+        pages[key] = data, {'url': url, 'sha256': hashlib.sha256(data).hexdigest(), 'bytes': len(data),
+                            'file': 'evidence/'+hashlib.sha256(data).hexdigest()+'.html',
+                            'observed_at': observed, 'http_status': 200, 'response_receipt': r}
+    return pages
+
+
+def collect(out, *, cache=None, snapshots=None):
+    pages = read_pages(cache, snapshots)
+    if [clean(n.text()) for n in Document(pages['index'][0].decode('utf-8')).root.walk() if n.tag == 'title'] != [TITLE]:
+        raise ValueError('Unexpected AAMAS 2025 introduction')
+    records, excluded = parse_contents(pages['contents'][0].decode('utf-8'), pages['contents'][1])
+    grouped = defaultdict(list)
+    for row in records: grouped[row['track']].append(row)
+    out = Path(out); out.mkdir(parents=True, exist_ok=True); files = []
+    with tempfile.TemporaryDirectory(prefix='.aamas2025-', dir=out) as temporary:
+        stage = Path(temporary); (stage/'evidence').mkdir()
+        for data, e in pages.values(): (stage/e['file']).write_bytes(data)
+        for track, group in sorted(grouped.items()):
+            p = stage/f'aamas-proceedings-2025-{track}.jsonl'
+            p.write_text(''.join(json.dumps(r, ensure_ascii=False)+'\n' for r in group))
+            m = {'source': 'aamas_proceedings', 'parser_version': 2, 'venue': 'AAMAS', 'year': 2025,
+                 'track': track, 'count': len(group), 'file': str(out/p.name), 'file_sha256': hashlib.sha256(p.read_bytes()).hexdigest(),
+                 'list_url': URLS['contents'], 'status': 'partial', 'missing_abstract_count': len(group),
+                 'publication_format_counts': dict(Counter(r['metadata']['publication_format'] for r in group)),
+                 'note': 'Complete published TOC partition; HTML abstracts absent, identities and submission-track scope require independent review.'}
+            p.with_suffix('.manifest.json').write_text(json.dumps(m, ensure_ascii=False, indent=2)+'\n'); files.append(m)
+        result = {'source': 'aamas_proceedings', 'year': 2025, 'finished_at': timestamp(),
+                  'status': 'collected_pending_identity_review', 'records': len(records), 'files': files,
+                  'excluded_publications': excluded, 'source_receipts': [e for _, e in pages.values()],
+                  'publication_format_counts': dict(Counter(r['metadata']['publication_format'] for r in records)),
+                  'database_written': False, 'pdfs_downloaded': False,
+                  'note': 'Neither absence from proceedings nor aggregate rejection/withdrawal counts identify an individual negative decision.'}
+        report = 'aamas-proceedings-2025.json'; (stage/report).write_text(json.dumps(result, ensure_ascii=False, indent=2)+'\n')
+        (out/'evidence').mkdir(exist_ok=True)
+        for p in (stage/'evidence').iterdir(): p.replace(out/'evidence'/p.name)
+        for p in stage.iterdir():
+            if p.is_file() and p.name != report: p.replace(out/p.name)
+        (stage/report).replace(out/report)
+    return result
@@ -6,6 +6,7 @@ import xml.etree.ElementTree as ET
 from pathlib import Path

 from paper_atlas.io.git_metadata import checkout_metadata
+from paper_atlas.collectors.acl_historical import event_scope, material_reason

 VENUES={'acl':'ACL','emnlp':'EMNLP','naacl':'NAACL','eacl':'EACL','coling':'COLING','conll':'CoNLL'}

@@ -29,7 +30,10 @@ def paper_id(collection,volume,number):
    return f'{collection}-{int(volume):0{width}d}{int(number):0{digits}d}'


-def volume_scope(meta,volume):
+def volume_scope(meta,volume,collection=None):
+    historical=event_scope(meta,volume,collection)
+    if historical:
+        return historical['canonical_venue'],historical['venue_aliases'],historical['track']
    codes=[text(n).lower() for n in meta.findall('venue')]
    book=text(meta.find('booktitle'))
    lower=book.lower()
@@ -59,7 +63,7 @@ def collect_xml(path,out,commit):
        vid=volume.attrib['id']
        meta=volume.find('meta')
        if meta is None:raise ValueError('Volume metadata missing')
-        scope=volume_scope(meta,vid)
+        scope=volume_scope(meta,vid,collection)
        key=f'{collection}-{vid}'
        if scope is None:
            results.append({'volume':key,'status':'excluded_by_venue','venue_codes':[text(n) for n in meta.findall('venue')],'booktitle':text(meta.find('booktitle')),'year':text(meta.find('year'))})
@@ -70,12 +74,18 @@ def collect_xml(path,out,commit):
        year=int(year_text)
        records=[]
        frontmatter=0
+        excluded_materials=[]
+        historical=event_scope(meta,vid,collection)
        for paper in volume.findall('paper'):
            pid=paper.attrib['id']
            if int(pid)==0:
                frontmatter+=1
                continue
            identifier=paper_id(collection,vid,pid)
+            reason=material_reason(collection,vid,paper)
+            if reason:
+                excluded_materials.append({'external_id':identifier,'title':text(paper.find('title')),'reason':reason})
+                continue
            doi=text(paper.find('doi'))
            if doi.startswith('10.18653/v1/') and doi.split('10.18653/v1/',1)[1].casefold()!=identifier.casefold():
                raise ValueError('Official DOI disagrees with constructed Anthology ID: '+identifier)
@@ -83,6 +93,11 @@ def collect_xml(path,out,commit):
            if not title:raise ValueError('Missing paper title: '+identifier)
            authors=[{'given':text(a.find('first')),'family':text(a.find('last'))} for a in paper.findall('author')]
            records.append({'source':'acl_anthology','external_id':identifier,'url':f'https://aclanthology.org/{identifier}/','title':title,'abstract':text(paper.find('abstract')) or None,'authors':authors,'identifiers':{'acl':identifier,'doi':doi},'venue':venue,'venue_aliases':aliases if len(aliases)>1 else [],'year':year,'track':track,'status':'accepted','raw_decision':'listed in authoritative ACL Anthology proceedings XML','metadata':{'collection':collection,'volume':vid,'booktitle':text(meta.find('booktitle')),'venue_codes':[text(n) for n in meta.findall('venue')],'pages':text(paper.find('pages')),'publisher':text(meta.find('publisher')),'metadata_commit':commit,'source_xml':f'https://github.com/acl-org/acl-anthology/blob/{commit}/data/xml/{collection}.xml'}})
+            if historical:
+                records[-1]['metadata']['historical_event_scope']=historical
+                if historical.get('publication_format'):
+                    records[-1]['metadata']['publication_format']=historical['publication_format']
+                    records[-1]['tags']=[{'namespace':'acl_anthology:publication_format','value':historical['publication_format']}]
        if not records:
            results.append({'volume':key,'status':'unavailable','reason':'no research paper elements after excluding front matter'})
            continue
@@ -93,7 +108,7 @@ def collect_xml(path,out,commit):
            for record in records:output.write(json.dumps(record,ensure_ascii=False)+'\n')
        part.replace(target)
        missing=sum(not row['abstract'] for row in records)
-        manifest={'source':'acl_anthology','venue':venue,'year':year,'track':track,'count':len(records),'xml_paper_elements':len(volume.findall('paper')),'excluded_frontmatter':frontmatter,'missing_abstract_count':missing,'list_url':f'https://aclanthology.org/{key}/','metadata_commit':commit,'file':str(target),'status':'partial' if missing else 'collected'}
+        manifest={'source':'acl_anthology','venue':venue,'year':year,'track':track,'count':len(records),'xml_paper_elements':len(volume.findall('paper')),'excluded_frontmatter':frontmatter,'excluded_materials':excluded_materials,'missing_abstract_count':missing,'list_url':f'https://aclanthology.org/{key}/','metadata_commit':commit,'file':str(target),'status':'partial' if missing else 'collected'}
        target.with_suffix('.manifest.json').write_text(json.dumps(manifest,indent=2))
        results.append(manifest)
        print(json.dumps(manifest),flush=True)
@@ -0,0 +1,76 @@
+"""Reviewed edition-specific scope; never alias entire source classifications."""
+
+HISTORY_URL = 'https://aclanthology.org/www.mt-archive.info/70/ACL-1982-Walker-1.pdf'
+NOTICE_URL = 'https://aclanthology.org/J77-4001.pdf'
+BOOK = 'Theoretical Issues in Natural Language Processing-2'
+MATERIAL_TITLE = 'List of questions suggested for consideration in each session'
+
+
+def text(node):
+    return ' '.join(''.join(node.itertext()).split()) if node is not None else ''
+
+
+def event_scope(meta, volume, collection):
+    if (collection, volume) == ('1963.earlymt', '1'):
+        return amtcl1963_scope(meta)
+    if (collection, volume) != ('T78', '1'):
+        return None
+    if ([text(n) for n in meta.findall('venue')] != ['tinlap']
+            or [text(n) for n in meta.findall('year')] != ['1978']
+            or [text(n) for n in meta.findall('booktitle')] != [BOOK]
+            or [(text(n.find('first')), text(n.find('last'))) for n in meta.findall('editor')]
+               != [('David L.', 'Waltz')]):
+        raise ValueError('Reviewed T78-1 scope metadata changed; independent review required')
+    return {
+        'collection': 'T78', 'volume': '1', 'original_venue': 'TINLAP',
+        'event_year': 1978, 'event_dates': ['1978-07-25', '1978-07-27'],
+        'relationship': 'ACL annual meeting took the form of TINLAP-2',
+        'canonical_venue': 'ACL+TINLAP', 'venue_aliases': ['ACL', 'TINLAP'],
+        'track': 'proceedings_unspecified',
+        'track_scope_note': 'Invited and submitted contributions are described at volume level; individual roles are not identified in XML.',
+        'evidence': [
+            {'url': NOTICE_URL, 'sha256': '14b5deb471c55afb07d789529b95d0462a2a1a77df3dfba7fa1f5d3ab75eb411',
+             'pdf_pages': [9, 10], 'printed_pages': ['9', '10'],
+             'scope': 'Contemporary annual meeting notice, not individual decision notes; ACM confirmation was pending in this notice.'},
+            {'url': HISTORY_URL, 'sha256': '35c21bd4983975bcaccbb94bd8ab91c9019fc190feec1791255407cf18a7e49a',
+             'pdf_pages': [3], 'printed_pages': ['91'],
+             'scope': 'Retrospective annual meeting table and TNLP abbreviation confirm the 1978 event association.'},
+        ],
+    }
+
+
+def amtcl1963_scope(meta):
+    expected = {
+        'venue': ['earlymt'], 'year': ['1963'],
+        'booktitle': ['Proceedings of the Annual meeting of the Association for Machine Translation and Computational Linguistics'],
+        'address': ['Denver, Colorado'], 'month': ['25-26 August'],
+    }
+    if any([text(n) for n in meta.findall(field)] != values
+           for field, values in expected.items()):
+        raise ValueError('Reviewed AMTCL 1963 scope metadata changed; independent review required')
+    return {
+        'collection': '1963.earlymt', 'volume': '1', 'original_venue': 'EarlyMT',
+        'original_event_name': 'Association for Machine Translation and Computational Linguistics',
+        'event_year': 1963, 'event_dates': ['1963-08-25', '1963-08-26'],
+        'relationship': 'Annual meeting under the former name of ACL; EarlyMT is the source classification, not a global ACL alias.',
+        'canonical_venue': 'ACL', 'venue_aliases': ['ACL', 'AMTCL'],
+        'track': 'proceedings_unspecified',
+        'track_scope_note': 'Published abstracts with named authors; individual submission tracks and original decision notes are not supplied.',
+        'publication_format': 'published_abstract',
+        'evidence': [
+            {'url': 'https://2023.aclweb.org/participants/about_acl/',
+             'sha256': '0f6bc260c2b34c9e2a51d971aeb43c3270a5b15b6db1d6d41e50669f316345f3',
+             'scope': 'Official ACL history identifies AMTCL as the former name, changed to ACL in 1968.'},
+            {'url': 'https://github.com/acl-org/acl-anthology/blob/a7be623d01c48c61405dec31c871121cdbed5a01/data/xml/1963.earlymt.xml',
+             'sha256': '4a4626d792ac507de0281cb358769f7246f4a0d6b6a9176ac02e61545c1635fc',
+             'scope': 'Official XML supplies the exact annual meeting title, dates, location and 34 contributions; hash is XML, not HTML.'},
+        ],
+    }
+
+
+def material_reason(collection, volume, paper):
+    if (collection, volume, paper.attrib.get('id')) != ('T78', '1', '36'):
+        return None
+    if text(paper.find('title')) != MATERIAL_TITLE or paper.findall('author'):
+        raise ValueError('Reviewed T78-1036 material changed; independent review required')
+    return 'Session question list, not an authored research contribution; no decision inferred.'
@@ -0,0 +1,225 @@
+"""Frozen official ICLR 2014–2016 lists; presentation evidence is not a track."""
+import argparse
+from collections import Counter, defaultdict
+import fcntl
+import hashlib
+import json
+from pathlib import Path
+import re
+import tempfile
+from urllib.parse import urlsplit
+
+from paper_atlas.db.submissions import timestamp
+from paper_atlas.io.html import Document, Node, clean
+from paper_atlas.io.http import fetch
+
+SOURCE = 'iclr_archive'
+PAGES = {
+    2014: ('https://iclr.cc/archive/2014/conference-proceedings/', {'main': 35}),
+    2015: ('https://iclr.cc/archive/www/doku.php%3Fid=iclr2015:accepted-main.html', {'main': 31, 'workshop': 74}),
+    2016: ('https://iclr.cc/archive/www/doku.php%3Fid=iclr2016:accepted-main.html', {'main': 80}),
+}
+SECTIONS = {
+    2015: {'Main Conference - Oral Presentations': ('main', 'oral'),
+           'Main Conference - Poster Presentations': ('main', 'poster'),
+           'Workshop Papers': ('workshop', 'workshop')},
+    2016: {'Oral Presentations': ('main', 'oral'), 'Poster Presentations': ('main', 'poster')},
+}
+
+
+def arxiv_link(url):
+    parsed = urlsplit(url)
+    if parsed.hostname != 'arxiv.org':
+        return None
+    match = re.fullmatch(r'/(?:abs|pdf)/(\d{4}\.\d{4,5})(v[1-9]\d*)?(?:\.pdf)?', parsed.path)
+    if (parsed.scheme not in ('http', 'https') or parsed.netloc != 'arxiv.org'
+            or parsed.query or parsed.fragment or not match):
+        raise ValueError('Unexpected official arXiv link')
+    return match[1], match[2] or None
+
+
+def without_links(node):
+    if node.tag in ('a', 'script', 'style'):
+        return ''
+    value = ''.join(without_links(n) if isinstance(n, Node) else n for n in node.children)
+    return ' '+value+' ' if node.tag in ('div', 'li', 'p', 'br') else value
+
+
+def authors(display, year):
+    parts = display.rstrip(';').split(';') if year == 2014 else re.split(r',\s*(?:and\s+)?|\s+and\s+', display)
+    names = [clean(p) for p in parts]
+    if not names or any(not name for name in names):
+        raise ValueError('Missing author or unsupported author delimiter')
+    return names
+
+
+def parse_page(html, year, evidence):
+    if year not in PAGES:
+        raise ValueError('Unreviewed ICLR archive year')
+    document = Document(html)
+    nodes = list(document.root.walk())
+    headings = [clean(n.text()) for n in nodes if n.tag in ('h1', 'h2', 'h3')]
+    if year == 2014:
+        if 'Conference Proceedings' not in headings or 'conference papers accepted' not in clean(document.root.text()):
+            raise ValueError('Unexpected ICLR 2014 accepted list')
+    elif (set(SECTIONS[year]) - set(headings)
+          or (year == 2016 and 'Accepted Papers (Conference Track)' not in headings)):
+        raise ValueError('Unexpected ICLR accepted sections')
+    entries = []
+    section = None
+    for index, node in enumerate(nodes):
+        if node.tag in ('h1', 'h2', 'h3'):
+            section = clean(node.text())
+        if year == 2014:
+            if node.tag != 'a' or not arxiv_link(node.attrs.get('href', '')):
+                continue
+            link = node
+            # This archived Google Sites page has an entry outside a paragraph.
+            # Authors can span several differently formatted italic spans.
+            # Read their whole paragraph, never only the first styled fragment.
+            display = None
+            for following in nodes[index+sum(1 for _ in link.walk()):]:
+                if following.tag == 'a' and arxiv_link(following.attrs.get('href', '')):
+                    break
+                if following.tag in ('i', 'em') or (following.tag == 'p' and any(
+                        n.tag in ('i', 'em') or 'font-style:italic' in n.attrs.get('style', '').replace(' ', '')
+                        for n in following.walk())):
+                    text = clean(following.text())
+                    if text:
+                        display = text
+                        break
+            if not display:
+                raise ValueError('Missing explicit ICLR 2014 author span')
+            track, presentation = 'main', 'conference'
+        else:
+            if node.tag != 'li':
+                continue
+            links = [n for n in node.walk() if n.tag == 'a' and arxiv_link(n.attrs.get('href', ''))]
+            if not links:
+                continue
+            if len(links) != 1 or section not in SECTIONS[year]:
+                raise ValueError('Ambiguous paper link or unknown accepted section')
+            link, = links
+            display = clean(without_links(node)).lstrip(',').strip()
+            track, presentation = SECTIONS[year][section]
+        arxiv, version = arxiv_link(link.attrs['href'])
+        title = clean(link.text())
+        if not title:
+            raise ValueError('Empty official title')
+        entries.append({'arxiv': arxiv, 'version': version, 'title': title,
+                        'authors': authors(display, year), 'authors_display': display,
+                        'paper_url': link.attrs['href'], 'track': track,
+                        'section': section, 'presentation': presentation})
+    # Check every arXiv occurrence, including links outside the expected layout.
+    occurrences = sum(n.tag == 'a' and bool(arxiv_link(n.attrs.get('href', ''))) for n in nodes)
+    if len(entries) != occurrences or not entries:
+        raise ValueError('Some official paper links were not parsed')
+    grouped = defaultdict(list)
+    for entry in entries:
+        grouped[(entry['track'], entry['arxiv'])].append(entry)
+    result = []
+    for (track, arxiv), variants in grouped.items():
+        first = variants[0]
+        if any((v['title'], v['authors']) != (first['title'], first['authors']) for v in variants):
+            raise ValueError('Repeated presentation has conflicting title or authors; review required')
+        presentations = sorted({v['presentation'] for v in variants})
+        result.append({'source': SOURCE, 'external_id': f'{year}/{track}/{arxiv}',
+                       'url': PAGES[year][0], 'title': first['title'], 'abstract': None,
+                       'authors': first['authors'], 'identifiers': {'arxiv': arxiv},
+                       'venue': 'ICLR', 'venue_aliases': [], 'year': year, 'track': track,
+                       'status': 'accepted', 'observed_at': evidence['observed_at'],
+                       'raw_decision': 'listed in official ICLR accepted '+track+' list',
+                       'metadata': {'source_evidence': evidence, 'source_variants': variants,
+                                    'presentations': presentations, 'decision_date_precision': 'not_supplied',
+                                    'identity_note': 'Official arXiv link, not title matching; validate canonical identity before import.',
+                                    'decision_evidence': 'Archived accepted list, not an original decision note.'},
+                       'tags': [{'namespace': SOURCE+':presentation', 'value': value} for value in presentations]})
+    return result
+
+
+def read_page(year, *, cache=None, snapshots=None):
+    if bool(cache) == bool(snapshots):
+        raise ValueError('Choose exactly one cache or frozen snapshot directory')
+    url = PAGES[year][0]
+    if snapshots:
+        root = Path(snapshots).resolve()
+        receipts = json.loads((root/'receipts.json').read_bytes())['receipts']
+        matches = [r for r in receipts if r['url'] == url]
+        if len(matches) != 1:
+            raise ValueError('Expected one exact official response receipt')
+        receipt, = matches
+        filename = Path(receipt['file'])
+        path = root/filename
+        if (filename.is_absolute() or '..' in filename.parts or not path.resolve().is_relative_to(root)
+                or receipt['status'] != 200):
+            raise ValueError('Unsafe or unsuccessful archive snapshot')
+        data = path.read_bytes()
+        if (len(data) > 8*1024**2 or len(data) != receipt['bytes']
+                or hashlib.sha256(data).hexdigest() != receipt['sha256']):
+            raise ValueError('Archive snapshot checksum or size differs')
+        observed = timestamp(receipt['finished_at'])
+    else:
+        data = fetch(url, cache, 8*1024**2, delay=1)
+        receipt = json.loads((Path(cache)/hashlib.sha256(url.encode()).hexdigest()).with_suffix('.json').read_bytes())
+        observed = timestamp(receipt['retrieved_at'])
+    evidence = {'url': url, 'sha256': hashlib.sha256(data).hexdigest(), 'bytes': len(data),
+                'file': 'evidence/'+hashlib.sha256(data).hexdigest()+'.html',
+                'observed_at': observed, 'http_status': 200, 'response_receipt': receipt}
+    return data, evidence
+
+
+def collect(year, out, *, cache=None, snapshots=None):
+    if year not in PAGES:
+        raise ValueError('Only independently reviewed ICLR archive years are supported')
+    out = Path(out)
+    out.mkdir(parents=True, exist_ok=True)
+    with (out/'iclr-archive.collect.lock').open('a') as lock:
+        fcntl.flock(lock, fcntl.LOCK_EX | fcntl.LOCK_NB)
+        report = out/f'iclr-archive-{year}.json'
+        if report.exists():
+            raise FileExistsError('Existing completed archive output must be inspected, not overwritten')
+        data, evidence = read_page(year, cache=cache, snapshots=snapshots)
+        records = parse_page(data.decode('utf-8'), year, evidence)
+        counts = Counter(r['track'] for r in records)
+        if dict(counts) != PAGES[year][1]:
+            raise ValueError('Distinct archive entries differ from reviewed inventory')
+        files = []
+        with tempfile.TemporaryDirectory(prefix='.iclr-archive-', dir=out) as temporary:
+            stage = Path(temporary)
+            (stage/'evidence').mkdir()
+            (stage/evidence['file']).write_bytes(data)
+            for track in sorted(counts):
+                group = [r for r in records if r['track'] == track]
+                path = stage/f'iclr-archive-{year}-{track}.jsonl'
+                path.write_text(''.join(json.dumps(r, ensure_ascii=False)+'\n' for r in group))
+                manifest = {'source': SOURCE, 'parser_version': 1, 'venue': 'ICLR', 'year': year,
+                            'track': track, 'count': len(group), 'file': str(out/path.name),
+                            'file_sha256': hashlib.sha256(path.read_bytes()).hexdigest(),
+                            'list_url': PAGES[year][0], 'status': 'partial', 'missing_abstract_count': len(group)}
+                path.with_suffix('.manifest.json').write_text(json.dumps(manifest, ensure_ascii=False, indent=2)+'\n')
+                files.append(manifest)
+            result = {'source': SOURCE, 'year': year, 'records': len(records), 'files': files,
+                      'source_evidence': evidence, 'database_written': False, 'pdfs_downloaded': False,
+                      'status': 'collected_pending_identity_review',
+                      'note': 'Accepted list only. Oral/poster are presentation evidence; absent papers are not rejected. '
+                              'ICLR2014 workshop and other unlisted tracks remain outside these reviewed pages.'}
+            (stage/report.name).write_text(json.dumps(result, ensure_ascii=False, indent=2)+'\n')
+            (out/'evidence').mkdir(exist_ok=True)
+            for path in (stage/'evidence').iterdir():
+                path.replace(out/'evidence'/path.name)
+            for path in stage.iterdir():
+                if path.is_file() and path.name != report.name:
+                    path.replace(out/path.name)
+            (stage/report.name).replace(report)
+        return result
+
+
+def main(argv=None, *, prog=None):
+    parser = argparse.ArgumentParser(prog=prog)
+    parser.add_argument('--year', type=int, choices=sorted(PAGES), required=True)
+    parser.add_argument('--out', required=True)
+    source = parser.add_mutually_exclusive_group(required=True)
+    source.add_argument('--cache')
+    source.add_argument('--snapshots')
+    args = parser.parse_args(argv)
+    print(json.dumps(collect(args.year, args.out, cache=args.cache, snapshots=args.snapshots), ensure_ascii=False))
@@ -0,0 +1,54 @@
+<HTML><HEAD><TITLE>AAMAS 2025: Proceedings of the 2025 International Conference on Autonomous Agents and Multiagent Systems</TITLE>
+<META http-equiv=Content-Type content="text/html; charset=utf-8"><LINK 
+href="forms/index_files/factemplate.css" type=text/css rel=stylesheet>
+<META content="MSHTML 6.00.2900.3562" name=GENERATOR></HEAD>
+<BODY bgcolor="#010101" bgproperties="fixed"><DIV style="TEXT-ALIGN: center" >
+<table width="724" border="0" align="center" cellpadding="0" cellspacing="0" valign="top">
+  <td colspan="2" align="center" valign="top" bgcolor="#FFFFFF"><img src="forms/index_files/aamas25.png" width="750" height="300" alt=""/></td>
+  </tr>
+  <tr bgcolor="#FFFFFF"> 
+    <td width="125" align="center" valign="top" bgcolor="#FFFFFF"><table width="125" border="1" align="center" cellpadding="1" cellspacing="1" bordercolor="#444444" style="margin-bottom: 2">
+        <tr> 
+          <td bgcolor="#FFFFFF"><p style="text-indent: 0; margin-left: 6; margin-right: 6; margin-top: 5; margin-bottom: 4" align="left"><A href="forms/index.htm">Main 
+              Page</A></p></td>
+        </tr>
+        <tr> 
+          <td bgcolor="#FFFFFF"><p style="text-indent: 0; margin-left: 6; margin-right: 6; margin-top: 5; margin-bottom: 4" align="left"><A href="forms/contents.htm">Table 
+              of Contents</A></p></td>
+        </tr>
+      </table></td>
+    <td width="599"><p style="text-indent: 0; margin-left: 12; margin-right: 14; margin-top: 12; margin-bottom: 12" align="left"><font color="#010101" size="4" face="Arial, Helvetica, sans-serif"><strong>Welcome 
+        to the Proceedings of AAMAS-2025</strong></font></p>
+      <p style="text-indent: 0; margin-left: 12; margin-right: 14; margin-top: 12; margin-bottom: 12;" align="left"><font color="#333333" size="2" face="Times New Roman, Times, serif">These 
+        are the proceedings of the 24th International Conference on Autonomous 
+        Agents and Multiagent Systems (AAMAS-2025). They are published by the 
+        International Foundation for Autonomous Agents and Multiagent Systems 
+    (IFAAMAS).</font></p></td>
+  </tr>
+  <tr bgcolor="#FFFFFF"> 
+    <td align="center" valign="middle" bgcolor="#FFFFFF">&nbsp;</td>
+    <td><table width="96%" border="0" cellspacing="2" cellpadding="2">
+        <tr align="left" valign="top"> 
+          <td> <p style="text-indent: 0; margin-left: 12; margin-right: 6; margin-top: 0; margin-bottom: 6;" align="left"><font color="#333333" face="Times New Roman, Times, serif"><font size="2"><strong>ISBN: 979-8-4007-1426-9</strong></font></font></p></td>
+          <td> <p style="text-indent: 0; margin-left: 12; margin-right: 6; margin-top: 0; margin-bottom: 6;" align="left"><font color="#333333" face="Times New Roman, Times, serif"></font></p></td>
+        </tr>
+        <tr> 
+          <td width="42%"><div align="left"><img src="forms/index_files/ifaamas.jpg" width="256" height="104"></div></td>
+          <td width="58%"><div align="center"><img src="forms/index_files/ACM-In-Cooperation_medium.jpg" width="285" height="86"></div></td>
+        </tr>
+      </table>
+      <p style="text-indent: 0; margin-left: 12; margin-right: 14; margin-top: 12; margin-bottom: 12;" align="left"><font color="#333333" size="2" face="Times New Roman, Times, serif">Copyright 
+        &copy; 2025 by International Foundation for Autonomous Agents and Multiagent 
+        Systems (IFAAMAS). Permission to make digital or hard copies of portions 
+        of this work for personal or classroom use is granted without fee provided 
+        that copies are not made or distributed for profit or commercial advantage 
+        and that copies bear this notice and the full citation on the first page. 
+        Copyright for components of this work owned by others than IFAAMAS must 
+        be honored. Abstracting with credit is permitted. To copy otherwise, to 
+        republish, to post on servers or to redistribute to lists, requires prior 
+    specific permission and/or a fee. </font></p></td>
+  </tr>
+</table>
+  <p>&nbsp;</p>
+
+</BODY></HTML>
@@ -0,0 +1,38 @@
+{
+  "status": "two_successful_official_metadata_pages_recovered_third403_not_retried",
+  "receipts": [
+    {
+      "file": "index.html",
+      "url": "https://www.ifaamas.org/Proceedings/aamas2025/",
+      "bytes": 4107,
+      "sha256": "850b7825c66d0147ac8a8bbf03a554590e4a9e8f4b5c576e043999bbd43317f4",
+      "at": "2026-11-09T12:31:49.347577+00:00",
+      "at_precision": "filesystem_write_time_recovered",
+      "http_status": 200,
+      "content_type": "text/html",
+      "receipt_recovered": true,
+      "scope": "Collectorasserted200/htmlbeforewrite;request3acceptedprogram403;noheaders/finalredirectURLpersisted;no refetch or rawbytesmutation."
+    },
+    {
+      "file": "contents.html",
+      "url": "https://www.ifaamas.org/Proceedings/aamas2025/forms/contents.htm",
+      "bytes": 370650,
+      "sha256": "732847b51972f34c721a49d8bd963a80a029a48f3ec08022c9a0d5ddc2945549",
+      "at": "2026-11-09T12:31:52.006642+00:00",
+      "at_precision": "filesystem_write_time_recovered",
+      "http_status": 200,
+      "content_type": "text/html",
+      "receipt_recovered": true,
+      "scope": "Collectorasserted200/htmlbeforewrite;request3acceptedprogram403;noheaders/finalredirectURLpersisted;no refetch or rawbytesmutation."
+    }
+  ],
+  "failed_url": "https://aamas2025.org/index.php/conference/program/accepted-papers/",
+  "not_attempted_url": "https://aamas2025.org/index.php/conference/calls/submission-instructions-main-technical-track/",
+  "original_error": {
+    "error": "HTTP Error 403: Forbidden",
+    "type": "HTTPError",
+    "downloaded": 2
+  },
+  "research_fulltext_downloaded": false,
+  "database_written": false
+}
@@ -0,0 +1,263 @@
+<?xml version='1.0' encoding='UTF-8'?>
+<collection id="1963.earlymt">
+  <volume id="1" ingest-date="2021-11-14" type="proceedings">
+    <!-- https://aclanthology.org/1963.earlymt-1/ -->
+    <meta>
+      <booktitle>Proceedings of the Annual meeting of the Association for Machine Translation and Computational Linguistics</booktitle>
+      <address>Denver, Colorado</address>
+      <month>25-26 August</month>
+      <year>1963</year>
+      <pdf hash="2ff28cb1"/>
+      <venue>earlymt</venue>
+    </meta>
+    <paper id="1">
+      <!-- https://aclanthology.org/1963.earlymt-1.1/ -->
+      <title>Necessity of introducing some information provided by transformational analysis into <fixed-case>MT</fixed-case> algorithms</title>
+      <author><first>Irena</first><last>Bellert</last></author>
+      <abstract>A few examples of ambiguous English constructions and their Polish equivalents are discussed in terms of the correlation between their respective phrase-marker representations and transformational analyses. It is shown by these examples that such an investigation can reveal interesting facts for MT, and therefore should be carried out for any pair of languages for which a given MT program is being constructed. If the phrase-marker of the English construction is set into one-to-one correspondence with the phrasemarker of the Polish equivalent construction, whatever particular transformational analysis of this construction is to be taken into account, then the ambiguous phrasemarker representation can be used as a syntactical model for MT algorithms with good results. If the phrase-marker of the English construction is set into one-to-many correspondence with the phrasemarkers of the Polish equivalents, according to the transformational analyses of this construction, then the ambiguous phrase-structure representation has to be resolved in terms of transformational analysis, for only then is it possible to assign the corresponding phrase structure representation to the Polish equivalents. A tentative scheme of syntactical recognition is provided for the multiply ambiguous adjectival construction in English1 (which proved to belong to the latter case) by means of introducing some information obtained from the transformational analysis of this construction.</abstract>
+      <bibkey>bellert-1963-necessity</bibkey>
+    </paper>
+    <paper id="2">
+      <!-- https://aclanthology.org/1963.earlymt-1.2/ -->
+      <title>The use of a random access device for dictionary lookup</title>
+      <author><first>Robert S.</first><last>Betz</last></author>
+      <author><first>Walter</first><last>Hoffman</last></author>
+      <abstract>The purpose of this paper will be to present a scheme to locate for single textual items and idioms in textual order their corresponding dictionary entries stored in an IBM 1301 random access mechanism. Textual items are considered to be 24 characters in length (left justified with following blanks). A dictionary entry consists of a 24 character Russian form, grammar information for the form and a set of translations for that form. Dictionary entries are packed into sequential tracks of the 1301. This paper will cover the method used for dictionary storage. The lookup for a textual item I first consists of a search for the first track that the dictionary entry E (if one exists) for I could be stored in. Once a track has been determined its contents are searched in core by a bisection convergence technique to find E. If E cannot be found, a “no entry” indication is made. If E is found a further search is made of the dictionary to find the longest sequence of text, starting with the first item I, that has a dictionary entry. The last such entry found is picked up. Included in the presentation will be examples of the dictionary lookup output for actual text.</abstract>
+      <bibkey>betz-hoffman-1963-use</bibkey>
+    </paper>
+    <paper id="3">
+      <!-- https://aclanthology.org/1963.earlymt-1.3/ -->
+      <title>Generative processes for <fixed-case>R</fixed-case>ussian impersonal sentences</title>
+      <author><first>C. G.</first><last>Borkowski</last></author>
+      <author><first>L. R.</first><last>Micklesen</last></author>
+      <abstract>Impersonal sentences of Russian are those traditionally construed to consist of predicates only. Ever since the first Russian grammar was compiled, they have continued to pose a problem for grammarians. This paper is intended to be a review and evaluation of all types of the so-called impersonal sentences in the Russian language. The investigation of these sentences has been conducted in terms of their relationships to basic (kernel) sentences. Our paper attempts to define the origin for such impersonal sentences, i.e., how such sentences might be derived within the framework of a generative grammar from a set of rules possessing maximal simplicity and maximal generative power. The long-range aim of this investigation involves the most efficient manipulation of such sentences in a recognition device for Russian-English MT.</abstract>
+      <bibkey>borkowski-micklesen-1963-generative</bibkey>
+    </paper>
+    <paper id="4">
+      <!-- https://aclanthology.org/1963.earlymt-1.4/ -->
+      <title>Concerning the role of sub-grammars in machine translation</title>
+      <author><first>Joyce M.</first><last>Brady</last></author>
+      <author><first>William B.</first><last>Estes</last></author>
+      <abstract>The comprehensive grammars being developed at the Linguistics Research Center of the University of Texas will be too large for easy access and manipulation in either experimental programs or practical translation. It is necessary, therefore, to devise some reliable method for selecting subsets of the grammar rules which will be reasonably adequate for a given purpose. Since the majority of the rules are dictionary rules, this problem is closely related both to the problem of constructing microglossaries and to the subsequent problem of choosing a particular microglossary suitable to a given text. Our current approach to this problem entails the construction of key word lists in the first stage of analysis which guide the computer in its choice of a previously constructed microglossary. Work to date indicates adaptations of this technique may not only contribute to the solution of storage and access problems but also facilitate analysis and simplify problems of semantic resolution.</abstract>
+      <bibkey>brady-estes-1963-concerning</bibkey>
+    </paper>
+    <paper id="5">
+      <!-- https://aclanthology.org/1963.earlymt-1.5/ -->
+      <title>Word-meaning and sentence-meaning</title>
+      <author><first>Elinor K.</first><last>Charney</last></author>
+      <abstract>A theory of semantics is presented which (1) defines the meanings of the most frequently occurring semantic morphemes (‘all’, ‘unless’, ‘only’, ‘if’, ‘not’, etc.), (2) explains their role, as semantically interdependent structural-constants, in giving rise to sentence-meanings, (3) suggests a possible approach to a sentenceby-sentence recognition program, and (4) offers a feasible method of coordinating among different language systems synonymous sentences whose grammatical features and structural-constants do not bear a one-to-one correspondence to one another. The theory applies only to morphemes that function as structural-constants and their interlocking relationships, denotative terms being treated as variables whose ranges alone have structural significance in sentencemeaning. The basic views underlying the theory are: In any given sentence, it is the particular configuration of structural-constants in combination with specific grammatical features which produces the sentencemeaning; the defined meaning of each individual structural-constant remains constant. The word-meanings of this type of morpheme, thus, must be carefully distinguished from the sentence-meanings that configuration of these morphemes produce. Sentence-synonymy is not based upon word-synonymy alone. Contrary to the popular view that the meanings of all of the individual words must be known before the sentencemeaning can be known, it is shown that one must comprehend the total configuration of structural-constants and syntactical features in a sentence in order to comprehend the correct sentence-meaning and that this understanding of the sentence as a whole must precede the determination of the correct semantic interpretation of these critical morphemes. In fact, the structural features that produce the sentence-meanings may restrict the possible meanings of even the denotative terms since a structural feature may demand, for example, a verbal rather than a noun phrase as an indispensable feature of the configuration. Two or more synonymous sentences whose denotative terms are everywhere the same but whose structural configurations are not isomorphic express the same fundamental sentence-meaning. The fundamental sentence-meanings can be explicitly formulated, and serve as the mapping functions to co-ordinate morphemically-unlike synonymous sentences within a language system or from one system to another. The research goal of the author is to establish empirically these translation rules that state formally the structural characteristics of the sentence configurations whose sentence-meanings, as wholes, are related as synonymous.</abstract>
+      <bibkey>charney-1963-word</bibkey>
+    </paper>
+    <paper id="6">
+      <!-- https://aclanthology.org/1963.earlymt-1.6/ -->
+      <title>Translating ordinary language into symbolic logic</title>
+      <author><first>Jared L.</first><last>Darlington</last></author>
+      <abstract>The paper describes a computer program, written in COMIT, for translating ordinary English into the notation of propositional logic and first-order functional logic. The program is designed to provide an ordinary language input to a COMIT program for the David-Putnam proof-procedure algorithm. The entire set of operations which are performed on an input sentence or argument are divided into three stages. In Stage I, an input sentence ‘S’, such as “The composer who wrote ‘Alcina’ wrote some operas in English,” is rewritten in a quasi-logical notation, “The X/A such that X/A is a composer and X/A wrote Alcina wrote some X/B such that X/B is an opera and X/B is in English.” The quasi-logical notation serves as an intermediate language between logic and ordinary English. In Stage II, S is translated into the logical notation of propositional functions and quantifiers, or of propositional logic, whichever is appropriate. In Stage III, S is run through the proof-procedure program and evaluated. (The sample sentence quoted is of course ‘invalid’, i.e. nontautological.) The COMIT program for Stage III is complete, that for Stage II is almost complete, and that for Stage I is incomplete. The paper describes the work done to date on the programs for Stages I and II.</abstract>
+      <bibkey>darlington-1963-translating</bibkey>
+    </paper>
+    <paper id="7">
+      <!-- https://aclanthology.org/1963.earlymt-1.7/ -->
+      <title>The graphic structure of word-breaking</title>
+      <author><first>J. L.</first><last>Dolby</last></author>
+      <author><first>H. L.</first><last>Resnikoff</last></author>
+      <abstract>In a recent paper1 the authors have shown that it is possible to determine the possible parts of speech of English words from an analysis of the written form. This determination depends upon the ability to determine the number of graphic syllables in the word. It is natural, then, to speculate as to the nature of graphic syllabification and the relation of this phenomenon to the practice of word-breaking in dictionaries and style manuals. It is not at all clear at the start that dictionary wordbreaking is subject to any fixed structure. In fact, certain forms cannot be broken uniquely in isolation since the dictionary provides different forms depending upon whether the word is used as a noun or a verb. However, it is shown in this paper that letter strings can be decomposed into 3 sets of roughly the same size in the following manner: in the first, strings are never broken in English words; in the second, the strings are always broken in English words; and in the third, both situations occur. Rules for breaking vowel strings are obtained by a study of the CVC forms. Breaks involving consonants can be determined by noting whether or not the consonant string occurs in penultimate position with the final c. The final e in compounds also serves to identify the forms that are generally split off from the rest of the word. A thorough analysis is made of the accuracy of the rules given when applied to the 12,000 words of the Government Printing Office Style Manual Supplement on word-breaking. Comparisons are also drawn between this source and several American dictionaries on the basis of a random sample of 500 words.</abstract>
+      <bibkey>dolby-resnikoff-1963-graphic</bibkey>
+    </paper>
+    <paper id="8">
+      <!-- https://aclanthology.org/1963.earlymt-1.8/ -->
+      <title>Writing of <fixed-case>C</fixed-case>hinese recognition grammar for machine translation</title>
+      <author><first>Ching-yi</first><last>Dougherty</last></author>
+      <abstract>Our approach to this problem is based on the stratificational grammar outlined and the procedures proposed by Dr. Sydney Lamb. How the theory and the procedures can be applied to written Chinese is briefly discussed. For the time being our research is limited to the particular kind of written Chinese found in chemical and biochemical journals. First the Chinese lexes are classified by detailed syntactical analysis, then binary grammar rules are constructed for joining two primary or constitute classes. How a more and more refined classification can eliminate one by one the ambiguity resulting from all possible constructions arising from juxtaposition of two distributional classes is discussed in detail.</abstract>
+      <bibkey>dougherty-1963-writing</bibkey>
+    </paper>
+    <paper id="9">
+      <!-- https://aclanthology.org/1963.earlymt-1.9/ -->
+      <title>The behavior of <fixed-case>E</fixed-case>nglish articles</title>
+      <author><first>H. P.</first><last>Edmundson</last></author>
+      <abstract>Machine translation has often been conceived as consisting of three steps: analysis of source-language sentence, transformation of analyzed pieces, and synthesis of target-language sentence. This paper is concerned with one aspect of the last step, namely, the rules of behavior of English articles. Since the classical definitions of definite and indefinite articles are operationally imprecise, proper mechanistic rules must be formulated in order to permit the automatic insertion or non-insertion of English articles. The rules discussed are of syntactic origin; however, note is also taken of their semantic aspects. This paper describes the methods used to derive these rules and offers ideas for further research.</abstract>
+      <bibkey>edmundson-1963-behavior</bibkey>
+    </paper>
+    <paper id="10">
+      <!-- https://aclanthology.org/1963.earlymt-1.10/ -->
+      <title>On representing syntactic structure</title>
+      <author><first>E. R.</first><last>Gammon</last></author>
+      <abstract>The idea of sentence depth of Yngve (A Model and an Hypothesis for Language Structure, Proc. Am. Phil. Soc., Vol. 104, No. 5, Oct. 1960) is extended to the notion of “distance” between constituents of a construction. The distance between constituents is defined as a weighted sum of the number of IC cuts separating them. Yngve’s depth is then a maximum distance from a sentence to any of its words. Various systems of weighting cuts are investigated. For example, in endocentric structures we may require that the distance from an attribute to the structure exceeds the distance from the head to the structure, and in exocentric structures that the distances from each constituent to the structure are equal. Representations of constructions are considered which preserve the distance between constituents. It is shown that it is impossible to represent some sentences in Euclidean space with exact distances, but a representation may be found if only relative order is preserved. If more general spaces are used then exact distances may be represented. It follows that for a wide class of sentence types, there is a weighting, and a space, in which the distance preserving representations are identical with the diagrams of traditional grammar.</abstract>
+      <bibkey>gammon-1963-representing</bibkey>
+    </paper>
+    <paper id="11">
+      <!-- https://aclanthology.org/1963.earlymt-1.11/ -->
+      <title>La traduction automatique et l’enseignement du russe</title>
+      <author><first>Yves</first><last>Gentilhomme</last></author>
+      <abstract>Les recherches effectuées depuis quelques années en vue de la Traduction Automatique ont conduit à des méthodes de travail et à des résultats intéressant la pédagogie des langues. Une expérience d’enseignement du russe a l’usage des scientifiques fondée sur ces données a été poursuivie pendant deux ans à Paris (Centre National de la Recherche Scientifique et Faculté des Sciences), et a abouti à la publication d’un manuel. Le present compte-rendu a pour objet de préciser les principes généraux utilisés, la réaction des étudiants et le rendement pédagogique obtenu. 1. Graphes morphologiques: Les mots d’une même famille. Notion de base. La double ramification. Les graphes abstraits. Les néologismes scientifiques. 2. Graphes syntaxiques: La double structure d’une phrase. Multiplicité des modèles. Point de vue psychologique. Notion de fonction. Continuité et discontinuité. 3. Les séparateurs: La segmentation d’une phrase. Le vocabulaire prioritaire. 4. Théorie de la valence: macro et microcontexte. Qu’est-ce-que “connaître un mot”? 5. Point de vue de l’étudiant; point de vue du traducteur humain; et point de vue de l’Enseignant.</abstract>
+      <bibkey>gentilhomme-1963-la</bibkey>
+    </paper>
+    <paper id="12">
+      <!-- https://aclanthology.org/1963.earlymt-1.12/ -->
+      <title>Word and context association by means of linear networks</title>
+      <author><first>Vincent E.</first><last>Giuliano</last></author>
+      <abstract>This paper is concerned with the use of electrical networks for the automatic recognition of statistical associations among words and contexts present in written text. A general mathematical theory is proposed for association by means of linear transformations, and it is shown that this theory can be realized through use of passive linear electrical networks. Several smallscale experimental associative networks have been built, and are briefly described in the paper; one such device will be demonstrated in the course of the oral presentation of the paper. Some of the devices generate measures of association among index terms used to characterize a document collection, and between the index terms and the documents themselves. Another uses syntactic proximity within sentences as a criterion for the generation of word association measures. Examples are given of associations produced by these network devices. It is conjectured that the networkproduced association measures reflect two distinct types of linguistic association—“synonymy” association which reflects similarity of meaning, and “contiguity” association which reflects real-world relationships among designata.</abstract>
+      <bibkey>giuliano-1963-word</bibkey>
+    </paper>
+    <paper id="13">
+      <!-- https://aclanthology.org/1963.earlymt-1.13/ -->
+      <title>A study of the combinatorial properties of <fixed-case>R</fixed-case>ussian nouns</title>
+      <author><first>Kenneth E.</first><last>Harper</last></author>
+      <abstract>A statistical study was made of the extent to which Russian nouns enter into certain kinds of syntactic combination. The basis of the study was a corpus of 180,000 running words of Russian physics text prepared for analysis by the Automatic Language Data Processing group at The Rand Corporation; for each sentence of text the syntactic dependency of each word had been previously coded. A data retrieval program was applied, showing for each noun in text the number of occurrences (a) with at least one genitive noun dependent, (b) with at least one adjective dependent, and (c) with either type of dependent. A listing of all nouns in text (64,026 occurrences of 2,993 nouns) was prepared, ordered by frequency, and showing counts for a, b, and c above. Separate listings were prepared, showing for each noun that occurred 50 times or more the probability P that it would be modified in each of these three ways; these listings were ordered on P. The data suggests, among others, the following conclusions: there is statistical significance in the variability with which nouns enter into the given combinations; the partial interchangeability of adjective and genitive noun modification is supported; a general correspondence exists between combinatorial groupings of nouns and morphological or semantic groupings (concrete nouns have low P for genitive complementation, abstract nouns have high P, etc); the use of words in a given field of discourse can be determined empirically (e.g., the use of deverbative nouns either to indicate a process or the result of a process). It is suggested that the distributional approach is a useful supplement to traditional syntactic and semantic classification schemes, and that it is of direct utility in automatic parsing programs.</abstract>
+      <bibkey>harper-1963-study</bibkey>
+    </paper>
+    <paper id="14">
+      <!-- https://aclanthology.org/1963.earlymt-1.14/ -->
+      <title>Connectability calculations, syntactic functions, and <fixed-case>R</fixed-case>ussian syntax</title>
+      <author><first>David G.</first><last>Hays</last></author>
+      <abstract>A program for sentence-structure determination can be divided into routines for analysis of word order and for testing the grammatical connectability of pairs of sentence members. The present paper describes a connectability-test routine that uses the technique called code matching. This technique requires elaborate descriptions of individual items, say the words or morphemes listed in a dictionary, but it avoids the use of large tables or complicated programs for testing connectability. Development of the technique also leads to a certain clarification of the linguistic concepts of function, exocentrism, and homography. In the present paper, a format for the description of Russian items is offered and a program for testing the connectability of pairs of Russian items is sketched. This system recognizes nine dominative functions: subjective; first, second, and third complementary; first, second, and third auxiliary; modifying; and predicative. The nature of a program for testing connectability with respect to coordinative functions (coordination, apposition, etc.) is suggested.</abstract>
+      <bibkey>hays-1963-connectability</bibkey>
+    </paper>
+    <paper id="15">
+      <!-- https://aclanthology.org/1963.earlymt-1.15/ -->
+      <title>Punctuation and automatic syntactic analysis</title>
+      <author><first>Lydia</first><last>Hirschberg</last></author>
+      <abstract>In this paper we discuss how algorithms for automatic analysis can take advantage of information carried by the punctuation marks. We neglect stylistic aspects of punctuation because they lack universality of usage and we restrict ourselves to those rules which any punctuation must observe in order to be intelligible. This involves a concept we call “coherence” of punctuation. In order to define “coherence”, we introduce two characteristics, which we prove to be mutually independent, namely “separating power” and “syntactic function”. The separating power is defined by three experimental laws expressing the fact that two punctuation marks of different separating power prevent to a different extent syntactic links from crossing them. These laws are defined independently of any particular grammatical character of the punctuation marks or of the attached grammatical syntagms. On the other hand, whichever grammatical system we choose, we may assimilate the punctuation marks to the ordinary words, to the extent that we can assign to them a known grammatical character and function, well defined in any particular context. They differ however from the other words by their large number of homographs and synonyms i.e. by the fact that almost every punctuation mark can occur with almost every grammatical value in each particular case, and in quite similar contexts. The syntactic functions, in general, and in particular those of the punctuation marks, can be ordered according to an arbitrary scale of decreasing “value” of syntactic links, where the “value” of a link is directly related to the number of syntactic conditions the links must satisfy. The law of coherence, then, shows that in a given context, a particular punctuation mark cannot indistinctly represent all its homographs, so that a certain number of assumptions about its syntactic nature and function can be discarded. This law can be stated as follows: “When moving from a punctuation mark to its immediate (left or right) neighbor in any text, the separating power cannot increase if the value of the syntactic function increases and vice-versa”. In addition we review two related topics, namely the stylistic character of punctuation and the necessity and existence of intrinsic criteria of grammatically, i.e. independent of punctuation. We propose such a criterion, and suggest a formalism related to the parenthesis free notation of logic.</abstract>
+      <bibkey>hirschberg-1963-punctuation</bibkey>
+    </paper>
+    <paper id="16">
+      <!-- https://aclanthology.org/1963.earlymt-1.16/ -->
+      <title>Application of decision tables to syntactic analysis</title>
+      <author><first>Walter</first><last>Hoffman</last></author>
+      <author><first>Amelia</first><last>Janiotis</last></author>
+      <author><first>Sidney</first><last>Simon</last></author>
+      <abstract>Decision tables have recently become an object of investigation as a possible means of improving problem formulation of data processing procedures. The initial emphasis for this new tool came from systems analysts who were primarily concerned with business data processing problems. The purpose of this paper is to investigate the suitability of decision tables as a means of expressing syntactic relations as an alternative to customary flow charting techniques. The history of decision tables will be briefly reviewed and several kinds of decision tables will be defined. As an example, parts of the predicative blocking routine developed at Wayne State University will be presented as formulated with the aid of decision tables. The aim of the predicative blocking routine is to group a predicative form together with its modal and temporal auxiliaries, infinitive complements, and negative particle, if any of these exist. The object of the search is to define such a syntactic block, but it may turn out instead that an infinitive phrase is defined or that a possible predicative form turns out to be an adverb.</abstract>
+      <bibkey>hoffman-etal-1963-application</bibkey>
+    </paper>
+    <paper id="17">
+      <!-- https://aclanthology.org/1963.earlymt-1.17/ -->
+      <title>Simultaneous computation of lexical and extralinguistic information measures in dialogue</title>
+      <author><first>Joseph</first><last>Jaffe</last></author>
+      <abstract>An approach to the study of information processing in verbal interaction is described. It compares patterns of two indices of dispersion in recorded dialogue. The lexical measure is the mean segmental type—token ratio, based on 25-word segments of the running conversation. It is computed from a key punched transcript of the dialogue without regard to the speaker of the words. The extralinguistic measure is the H statistic, computed from the temporal pattern of the interaction. The latter is prepared from a two-channel tape recording by a special analogue to digital converter (AVTA system) which key punches the state of the vocal transaction 200 times per minute. Probabilities of the four possible states (either A or B speaking, neither speaking, both speaking) are the basis for the computation. All analyses are done on the IBM 7090. The methodology is part of an investigation of information processing in dyadic systems, aimed toward the reclassification of pathological communication.</abstract>
+      <bibkey>jaffe-1963-simultaneous</bibkey>
+    </paper>
+    <paper id="18">
+      <!-- https://aclanthology.org/1963.earlymt-1.18/ -->
+      <title>Design of a generalized information system</title>
+      <author><first>Ronald W.</first><last>Jonas</last></author>
+      <abstract>While mechanical translation research involves the design of a computer system which simulates language processes, there is the associated problem of collecting the language data which are to be used in translation. Because large quantities of information will be needed, the computer may be useful for data accumulation and verification. A generalized information system should be able to accept the many types of data which a linguist encodes. A suitable means of communication between the linguist and the system has to be established. This may be achieved with a central input, called Linguistic Requests, and a central output, called Information Displays. The requests should be coordinated so that all possible inputs to the system are compatible, and the displays should be composed by the system such that they are clearly understandable. An information system should be interpretive of the linguist’s needs by allowing him to program the data manipulation. The key to such a scheme is that the linguist be permitted to classify his data freely and to retrieve it as he chooses. He should have at his disposal selecting, sorting, and displaying functions with which he can verify data, select data for introduction to a mechanical translation system, and perform other activities necessary in his research. Such an information system has been designed at the Linguistics Research Center of The University of Texas.</abstract>
+      <bibkey>jonas-1963-design</bibkey>
+    </paper>
+    <paper id="19">
+      <!-- https://aclanthology.org/1963.earlymt-1.19/ -->
+      <title>Some experiments performed with an automatic paraphraser</title>
+      <author><first>Sheldon</first><last>Klein</last></author>
+      <abstract>The automatic paraphrasing system used in the experiments described herein consisted of a phrase structure, grammatically correct nonsense generator coupled with a monitoring system that required the dependency relations of the sentence in production to be in harmony with those of a source text. The output sentences also appeared to be logically consistent with the content of that source. Dependency was treated as a binary relation, transitive except across most verbs and prepositions. Five experiments in paraphrasing were performed with this basic system. The first attempted to paraphrase without the operation of the dependency monitoring system, yielding grammatically correct nonsense. The second experiment included the operation of the monitoring system and yielded logically consistent paraphrases of the source text. The third and fourth experiments demanded that the monitoring system permit the production of only those sentences whose dependency relations were non-existent in the source text. While these latter outputs were seemingly nonsensical, they bore a special logical relationship to the source. The fifth experiment demanded that the monitoring system permit the production of sentences whose dependency relations were the converse of those in the source. This restriction was equivalent to turning the dependency tree of the source text upside down. The output of this experiment consisted only of kernel type sentences which, if read backwards, were logically consistent with the source. The results of these experiments determine some formal properties of dependency and engender some comments about the role of dependency in phrase structure and transformational models of language.</abstract>
+      <bibkey>klein-1963-experiments</bibkey>
+    </paper>
+    <paper id="20">
+      <!-- https://aclanthology.org/1963.earlymt-1.20/ -->
+      <title>Interlingual correspondence at the syntactic level</title>
+      <author><first>Edward S.</first><last>Klima</last></author>
+      <abstract>The paper will investigate a few major construction types in several related European languages: relative clauses, attributive phrases, and certain instances of coordinate conjunction involving these constructions. In each of the languages independently, the constructions will be described as resulting from syntactic mechanisms further analyzable into chains of partially ordered operations on more basic structures. Pairs of sentences equivalent in two languages will be examined. Sentences will be considered equivalent if they are acceptable translations of one another. The examples used will, in fact, be drawn primarily from standard translations of scholarly and literary prose. Equivalence between whole sentences can be further analyzed, as will be shown, into general equivalence 1) between the chains of operations describing the constructions and 2) between certain elements (e.g., lexical items) in the more basic underlying structures. It will be seen that superficial differences in the ultimate shape of certain translation pairs can be accounted for as the result of minor differences in the particular operations involved or in the basic underlying structure. We shall examine two languages (e.g., French and German) in which attributive phrase formation and relative clause formation on the whole correspond and in which, in a more or less abstract way, the rules of relative clause formation are included as intermediate links in the chain of operations describing attributive phrases. The fact that in particular cases a relative clause in the one language corresponds to an attributive phrase in the other will be found to result from, e.g., differences in the choice of perfect auxiliary in the two languages.</abstract>
+      <bibkey>klima-1963-interlingual</bibkey>
+    </paper>
+    <paper id="21">
+      <!-- https://aclanthology.org/1963.earlymt-1.21/ -->
+      <title>Sentence structure diagrams</title>
+      <author><first>Susumu</first><last>Kuno</last></author>
+      <abstract>A system for automatically producing a sentence structure diagram for each analysis of a given sentence has been added to the program of the multiple-path syntactic analyzer. A structure code, consisting of a series of structure symbols or phrase markers that identify the successive higher-order structures to which the word in question belongs, is assigned to each word of the sentence. The set of structure codes for the words of a given sentence is equivalent to an explicit tree diagram of the sentence structure, but more compact and easier to lay out on conventional printers. The diagramming system makes some experimental assumptions about the dependencies of certain structures upon higher-level structures. All the major syntactic components of a sentence (i.e., subject, verb, object, complement, period, or question mark) are represented in the current system as occurring on the same level, all being dependent on the topmost level, “sentence”. A floating structure such as a prepositional phrase or adverbial phrase or clause, whose dependency is not determined in the analyzer, is represented as depending upon the nearest preceding structure modifiable by such a floating structure. Different assumptions as to structural dependencies would yield different diagrams without requiring modification on the main flow of the diagramming program. The diagrams thus obtained contribute greatly to the rapid and accurate evaluation of the analysis results, and they are also useful for obtaining basic syntactic patterns of analyzed structures, and for detecting the head of each identified structure.</abstract>
+      <bibkey>kuno-1963-sentence</bibkey>
+    </paper>
+    <paper id="22">
+      <!-- https://aclanthology.org/1963.earlymt-1.22/ -->
+      <title>Linguistic structure and machine translation</title>
+      <author><first>Sydney M.</first><last>Lamb</last></author>
+      <abstract>If one understands the nature of linguistic structure, one will know what design features an adequate machine translation system must have. To put it the other way around, it is futile to attempt the construction of a machine translation system without a knowledge of what the structure of language is like. This principle means that if someone wants to construct a machine translation system, the most important thing he must do is to understand the structure of language. Any MT system, whether by conscious intention on the part of its creators or not, is based upon some view of the nature of linguistic structure. By making explicit the underlying theory for various MT systems which have been proposed we can determine whether or not they are adequate. Similarly, by observing linguistic phenomena we can determine what properties an adequate theory of language must have, and such determination will show what features an MT system must have in order to be adequate. It can be shown that some of the approaches to MT now being pursued must necessarily fail because their underlying linguistic theories are inadequate to account for various well-known linguistic phenomena.</abstract>
+      <bibkey>lamb-1963-linguistic</bibkey>
+    </paper>
+    <paper id="23">
+      <!-- https://aclanthology.org/1963.earlymt-1.23/ -->
+      <title>On redundancy in artificial languages</title>
+      <author><first>W. P.</first><last>Lehmann</last></author>
+      <abstract>Artificial languages are one concern of work in computational linguistics, if only as a mnemonic device for interlinguas which will be developed. Even if it does not gain wider use, the structure of an artificial language is of general interest. In contrast to the artificial languages which have been widely proposed, linguistic principles underlying a welldesigned artificial language and its usefulness are wellestablished, particularly through Trubetzkoy’s article, TCLP 8.5-21. which indicates phonological limitations for such a language. Since Trubetzkoy’s specifications yield a total of approximately 11,000 morphemes, if an artificial language incorporated the degree of redundancy found in natural languages it would be severely handicapped by the size of its lexicon. The paper discusses the problem particularly with regard to suprasegmentals, which Trubetzkoy almost entirely ignored.</abstract>
+      <bibkey>lehmann-1963-redundancy</bibkey>
+    </paper>
+    <paper id="24">
+      <!-- https://aclanthology.org/1963.earlymt-1.24/ -->
+      <title>A procedure for automatic sentence structure analysis</title>
+      <author><first>D.</first><last>Lieberman</last></author>
+      <abstract>The two main considerations in the design of this procedure were the economical recognition and representation of multiple readings of syntactically ambiguous sentences, and general applicability to “all” languages (English, Russian, Chinese). The following features will be discussed: types of structural descriptions, form of linguistic rules, use of linguistic heuristics to achieve economical multiple analyses, application to linguistic research and application to production MT systems. Also, the relation between this procedure and other existing sentence analysis procedures will be discussed.</abstract>
+      <bibkey>lieberman-1963-procedure</bibkey>
+    </paper>
+    <paper id="25">
+      <!-- https://aclanthology.org/1963.earlymt-1.25/ -->
+      <title>An algorithm for the translation of <fixed-case>R</fixed-case>ussian inorganic-chemistry terms</title>
+      <author><first>L. R.</first><last>Micklesen</last></author>
+      <author><first>P. H.</first><last>Smith</last></author>
+      <abstract>An algorithm has been devised, and a computer program written, to translate certain recurring types of inorganic-chemistry terms from Russian to English. The terms arc all noun-phrases, and several different types of such phrases have been included in the program. Examples are: AZOTNONATRIEVA4 SOL6 sodium nitrate SOL6 ZAKISI/OKISI JELEZA ferrous/ferric salt ZAKISNA4 OKISNA4 SOL6 JELEZA GIDRAT ZAKISI/OKISI JELEZA ferrous/ferric salt etc., where the stems underlined may be replaced by any of a number of other stems (up to 65 in some positions) in the particular type. Translation of each type encounters problems common to almost all the types: (1) The Russian noun is translated as an English adjective, while the noun of the resulting English phrase is found among the modifiers of the Russian noun. (2) The Russian noun (English adjective) may be a metal with more than one valence state, the state indicated (if at all) by the modifiers. (3) The number of the resulting English noun-phrase is determined by some member of the Russian phrase other than the noun. (4) The phrase elements may occur compounded in the chemical phrase but free in other contexts, and dictionary storage must provide for this. The program permits translation of conjoined phrase elements as well. The paper also includes an investigation into the deeper grammatical implications of this type of chemical nomenclature, and some excursions into the semantic correlations involved.</abstract>
+      <bibkey>micklesen-smith-1963-algorithm</bibkey>
+    </paper>
+    <paper id="26">
+      <!-- https://aclanthology.org/1963.earlymt-1.26/ -->
+      <title>The application of table processing concepts to the Sakai translation technique</title>
+      <author><first>A.</first><last>Opler</last></author>
+      <author><first>R.</first><last>Silverstone</last></author>
+      <author><first>Y.</first><last>Saleh</last></author>
+      <author><first>M.</first><last>Hildebran</last></author>
+      <author><first>I.</first><last>Slutzky</last></author>
+      <abstract>In 1961, I. Sakai described a new technique for the mechanical translation of languages. The method utilizes large tables which contain the syntactic rules of the source and target languages. As part of a study of the AN/GSQ-16 Lexical Processing Machine, a modification of the Sakai method was developed. Five of six planned table scanning phases were implemented and tested. Our translation system (1) converts input text to syntactic and semantic codes with a dictionary scan, (2) clears syntactic ambiguities where resolution by adjacent words is effective, (3) resolves residual syntactic ambiguities by determining the longest meaningful semantic unit, (4) reorders word sequence according to the rules of the target language and (5) produces the final target language translation. French to English was the source-target pair selected for the study. An Input Dictionary of 3,000 French stems was prepared and 17,000 entries comprised the Input Product Table (allowable syntactic combinations ). Since Sakai was working with highly dissimilar languages, he found it necessary to use an intermediate language. Because of the structural similarity between French and English, we found an intermediate language was unnecessary. The method proved straightforward to implement using the table lookup logic of the Lexical Processor. The translation was actually performed on an IBM 1401 which we programmed to simulate the concept of the AN/GSQ-16 Lexical Processor. In our implementation magnetic tapes replaced the photoscopic storage disk.</abstract>
+      <bibkey>opler-etal-1963-application</bibkey>
+    </paper>
+    <paper id="27">
+      <!-- https://aclanthology.org/1963.earlymt-1.27/ -->
+      <title><fixed-case>S</fixed-case>lavic languages—comparative morphosyntactic research</title>
+      <author><first>Milos</first><last>Pacak</last></author>
+      <abstract>An appropriate goal for present-day linguistics is the development of a general theory of relations between languages. One necessary requirement in the development of such a theory is the identification and classification of inflected forms in terms of their morphosyntactic properties in a set of presumably related languages. According to Sapir, “all languages differ from one another, but certain ones differ far more than others”. As for the Slavic languages he might well have said that they are all alike, but some are more alike than others. The similarities stemming from their common origin and from subsequent parallel development enable us to group them into a number of more or less homogeneous types. The experimental comparative research at The Georgetown University was focused on a group of four Slavic languages, namely, Russian, Czech, Polish and Serbocroatian. The first step in the comparative procedure here described is the morphosyntactic analysis of each of the four languages individually. The analysis should be based on the complementary distribution of inflectional morphemes. The properties whose distribution must be determined are: 1) the graphemic shape of the inflectional morphemes, 2) the establishment of distributional classes and subclasses of stem morphemes and (on the basis of 1 and 2), 3) the morphosyntactic function of inflectional morphemes which is determined by the distributional subclass of the stem morpheme. f(x,y)-l, where x is the distributional subclass of the stem morpheme (which is a constant) and y is the given inflectional morpheme (which is a free variable). On the basis of this preliminary analysis the patterns of absolute equivalence, partial equivalence, and absolute difference can be established for each class of inflected forms in each language under study. Once this has been accomplished, the results can be used in order to determine the extent of distributional equivalences among the individual languages. The applicability of this procedure was tested on the class of adjectivals. Within the frame of adjectivals the following morphosyntactic properties were analyzed within each language languages: 1) the category of gender, 2) the category of animateness, 3) the category of case and murder. The product of this comparative analysis is a set of formation rules which embody a system for the identification of the inflected forms. The detailed result will be presented in an additional report.</abstract>
+      <bibkey>pacak-1963-slavic</bibkey>
+    </paper>
+    <paper id="28">
+      <!-- https://aclanthology.org/1963.earlymt-1.28/ -->
+      <title>Types of language hierarchy</title>
+      <author><first>E. D.</first><last>Pendergraft</last></author>
+      <abstract>Various relations lead to hierarchical systems of linguistic description. This paper considers briefly a typology of descriptive metalanguages based on such relations and sketches possible consequences for computational linguistics. Its scope is accordingly limited to metalanguages having operational interpretations which specify individual linguistic processes and structural interpretations which specify language data of individual languages. Immediate-constituent, context-free metalanguages are used to illustrate hierarchical types.</abstract>
+      <bibkey>pendergraft-1963-types</bibkey>
+    </paper>
+    <paper id="29">
+      <!-- https://aclanthology.org/1963.earlymt-1.29/ -->
+      <title>Path economization in exhaustive left-to-right syntactic analysis</title>
+      <author><first>Warren J.</first><last>Plath</last></author>
+      <abstract>In exhaustive left-to-right syntactic analysis using the predictive approach, each path of syntactic connection which originates at the beginning of a sentence must be followed until it is clear whether or not it will lead to the production of a well-formed analysis. The original scheme of following each path until it terminates either in an analysis or in a grammatical inconsistency has been considerably improved through the. incorporation of two path-testing techniques. Using the first technique, the program abandons a path as unproductive whenever a situation is detected where the prediction pool contains more predictions of a given type than can possibly be fulfilled by the remaining words in the sentence. Employment of the second technique, which is based on periodic comparison of the current prediction pool with pools formed on earlier productive paths, eliminates repeated analysis of identical right-hand segments which belong to distinct paths. Taken together, the two path-testing procedures frequently enable the program to terminate the processing of a path well before its end has been reached. For most sentences, this means a considerable reduction in the total path length traversed, accompanied by a corresponding increase in the speed of analysis. Comparison of runs performed using both versions of the program indicates that employment of the new techniques reduces the average running time per sentence to less than one-fifth of its former value.</abstract>
+      <bibkey>plath-1963-path</bibkey>
+    </paper>
+    <paper id="30">
+      <!-- https://aclanthology.org/1963.earlymt-1.30/ -->
+      <title>A computer representation for semantic information</title>
+      <author><first>Bertram</first><last>Raphael</last></author>
+      <abstract>This paper deals with the problem of representing in a useful form, within a digital computer, the information content of statements in natural language. The model proposed consists of words and list-structure associations between words. Statements in simple English are thought of as describing relations between objects in the real world. Sentences are analyzed by matching them against members of a list of formats, each of which determines a unique relation. These relations are stored on description-lists associated with those words which denote objects (or sets of objects). A LISP computer program uses this model in the context of a simple question-answering system. Functions are provided which may grow, search, and modify this model. Formats and functions dealing with set-relations, part-whole and numeric relations, and left-toright spatial relations have been included in the system, which is being expanded to handle other types of relations. All functions which operate on the model report information concerning their actions to the programmer, so that the applicability and limitations of this kind of model may more easily be evaluated.</abstract>
+      <bibkey>raphael-1963-computer</bibkey>
+    </paper>
+    <paper id="31">
+      <!-- https://aclanthology.org/1963.earlymt-1.31/ -->
+      <title>Specifications for generative grammars used in language data processing</title>
+      <author><first>Robert</first><last>Tabory</last></author>
+      <abstract>It becomes more and more evident that successful pragmatics (i.e. automatic recognition and production procedures for sentences) cannot be performed without previously written generative grammars for the languages involved, using an underlying meta-theoretical framework proposed by the present school of mathematical linguistics. Two aspects of grammar writing are examined: 1. A taxonomy over the non-terminal vocabulary, using a subscripting system for signs and fitting into the more general string taxonomy of phrase structure components. The resulting more complex lexical organization is studied. 2. A command syntax for phrase structure components limiting the full, not necessarily needed generative power of these grammars. The proposed restrictions correspond to a priori linguistic intuition. Applicational order and location of the rules is studied. Finally, the recognitional power and generative capacity of a computer are examined, the machine being structured according to a Newell-Shaw-Simon list system. It is well known that pushdown stores are particular cases of list structures, that context-free grammars are particular cases of phrase structure grammars and that pushdown stores are the generative devices for context-free grammars.</abstract>
+      <bibkey>tabory-1963-specifications</bibkey>
+    </paper>
+    <paper id="32">
+      <!-- https://aclanthology.org/1963.earlymt-1.32/ -->
+      <title>Collecting linguistic data for the grammar of a language</title>
+      <author><first>Wayne</first><last>Tosh</last></author>
+      <abstract>Establishing the grammatical description of a language is one of the major tasks facing the technician in machine translation. Another is that of creating the system of programs with which to carry out the translation process. The Linguistics Research Center of The University of Texas recognizes the advantages in maintaining the specialties of linguistic research and computer programming as two separate areas of endeavor. We regard the linguistic task as a problem in convergence. We do not expect ever to have a final description of a language (except theoretically for a given point in the history of that language). We do expect, however, to begin with almost immediate application of the very first grammatical description. We shall make repeated revisions of the grammar as we learn how to make it approximate better the language text fed into the computer. The grammatical description of any one language is based primarily on specific text evidence. We are not attempting to describe “the language”. We are, however, attempting to make descriptive decisions sufficiently general that new text evidence does not require extensive revision of earlier descriptions. Corpora selected for description are chosen so as to have similar texts within the same scientific discipline for the several languages. Tree diagrams are drawn for each sentence in detail. The diagrams are inspected for consistency before corresponding phrase-structure rules are compiled in the computer. The grammar is then verified in the computer system and revised as necessary.</abstract>
+      <bibkey>tosh-1963-collecting</bibkey>
+    </paper>
+    <paper id="33">
+      <!-- https://aclanthology.org/1963.earlymt-1.33/ -->
+      <title>Derivational suffixes in <fixed-case>R</fixed-case>ussian general vocabulary and in chemical nomenclature</title>
+      <author><first>John H.</first><last>Wahlgren</last></author>
+      <abstract>A grammar based upon a conventional morphemic analysis of Russian will have a rather large inventory of derivational suffixes. A relatively small number of these recur with sufficient generality to acquire lexemic status (i.e., to be what is usually termed “productive”). Names of chemical substances in Russian may likewise be analyzed as combinations of roots or stems with derivational affixes, in particular, suffixes. The number of productive suffixes in the chemical nomenclature is considerably larger than in the general vocabulary. These suffixes derive from adoption into Russian of an international system of chemical nomenclature. A grammar of this system is basically independent of any grammar of Russian. It must, however, be consistently incorporated into the grammar and dictionary which are to serve in a machine translation system for texts in the source language containing chemical names. Grammatical analysis of chemical suffixes and connected study of general Russian derivational suffixes has raised certain practical problems and theoretical questions concerning the nature of derivation. On the practical side, where a complex and highly productive system is involved, effective means of detecting and dealing with homography have required development. Theoretical consideration has been given to the question of grammaticality in chemical names and to problems of sememic analysis and classification of root and stem lexemes into tactic classes on the basis of cooccurrence with derivational suffixes.</abstract>
+      <bibkey>wahlgren-1963-derivational</bibkey>
+    </paper>
+    <paper id="34">
+      <!-- https://aclanthology.org/1963.earlymt-1.34/ -->
+      <title>On the order of clauses</title>
+      <author><first>Victor H.</first><last>Yngve</last></author>
+      <abstract>We used to think that the output of a translation machine would be stylistically inelegant, but this would be tolerable if only the message got across. We now find that getting the message across accurately is difficult, but we may be able to have stylistic elegance in the output since much of style reflects depth phenomena and thus is systematic. As an example, the order of the clauses in many twoclause sentences can be reversed without a change of meaning, but the same is not normally true of sentences with more than two clauses. The meaning usually changes when the clause order is changed. Equivalently, there appear to be severe restrictions on clause order for any given meaning. These restrictions appear to follow from depth considerations. The idea is being investigated that there is a normal depth-related clause order and any deviations from this order must be signalled by special syntactic or semantic devices. The nature of these devices is being explored. When translating multi-clause sentences, there may be trouble due to the fact that the clause types of the two languages are not exactly parallel. Therefore the list of allowed and preferred clause orders in the two languages will not be equivalent and the special syntactic and semantic devices available to signal deviations from the normal order will be different. Thus one would predict that multi-clause sentences in language A often have to be split into two or more sentences when translated into language B, while at the same time multi-clause sentences in language B will often have to be broken into two or more sentences when translating into language A.</abstract>
+      <bibkey>yngve-1963-order</bibkey>
+    </paper>
+  </volume>
+</collection>
@@ -0,0 +1,273 @@
+<?xml version='1.0' encoding='UTF-8'?>
+<collection id="T78">
+  <volume id="1" type="proceedings">
+    <!-- https://aclanthology.org/T78-1/ -->
+    <meta>
+      <booktitle>Theoretical Issues in Natural Language <fixed-case>P</fixed-case>rocessing-2</booktitle>
+      <editor id="david-l-waltz"><first>David L.</first><last>Waltz</last></editor>
+      <venue>tinlap</venue>
+      <year>1978</year>
+    </meta>
+    <frontmatter>
+      <!-- https://aclanthology.org/T78-1000/ -->
+      <pdf hash="fb6164ad"/>
+      <bibkey>tinlap-1978-theoretical</bibkey>
+    </frontmatter>
+    <paper id="1">
+      <!-- https://aclanthology.org/T78-1001/ -->
+      <title>Testing The Psychological Reality of a Representational Model</title>
+      <author><first>Dedre</first><last>Gentner</last></author>
+      <pdf hash="f723a956"/>
+      <bibkey>gentner-1978-testing-psychological</bibkey>
+    </paper>
+    <paper id="2">
+      <!-- https://aclanthology.org/T78-1002/ -->
+      <title>What Makes Something “Ad Hoc”</title>
+      <author id="roger-c-schank"><first>Roger C.</first><last>Schank</last></author>
+      <pdf hash="636853f9"/>
+      <bibkey>schank-1978-makes-something</bibkey>
+    </paper>
+    <paper id="3">
+      <!-- https://aclanthology.org/T78-1003/ -->
+      <title>The Relation of Grammar to Cognition–a Synopsis</title>
+      <author><first>Leonard</first><last>Talmy</last></author>
+      <pdf hash="101c67dc"/>
+      <bibkey>talmy-1978-relation-grammar</bibkey>
+    </paper>
+    <paper id="4">
+      <!-- https://aclanthology.org/T78-1004/ -->
+      <title>On primitives, prototypes, and other semantic anomalies</title>
+      <author><first>Terry</first><last>Winograd</last></author>
+      <pdf hash="f8e47023"/>
+      <bibkey>winograd-1978-primitives-prototypes</bibkey>
+    </paper>
+    <paper id="5">
+      <!-- https://aclanthology.org/T78-1005/ -->
+      <title>Taxonomic Lattice Structures for Situation Recognition</title>
+      <author id="william-a-woods"><first>William A.</first><last>Woods</last></author>
+      <pdf hash="92957ddf"/>
+      <bibkey>woods-1978-taxonomic-lattice</bibkey>
+    </paper>
+    <paper id="6">
+      <!-- https://aclanthology.org/T78-1006/ -->
+      <title>Description Formation and Discourse Model Synthesis</title>
+      <author id="bonnie-webber"><first>Bonnie Lynn</first><last>Webber</last></author>
+      <pdf hash="ffded1d2"/>
+      <bibkey>webber-1978-description-formation</bibkey>
+    </paper>
+    <paper id="7">
+      <!-- https://aclanthology.org/T78-1007/ -->
+      <title>The Processing of Referring Expressions within a Semantic Network</title>
+      <author><first>John R.</first><last>Anderson</last></author>
+      <pdf hash="8bcc083a"/>
+      <bibkey>anderson-1978-processing-referring</bibkey>
+    </paper>
+    <paper id="8">
+      <!-- https://aclanthology.org/T78-1008/ -->
+      <title>Reference Diaries</title>
+      <author><first>Herbert H.</first><last>Clark</last></author>
+      <author><first>Catherine</first><last>Marshall</last></author>
+      <pdf hash="d06b7ea2"/>
+      <bibkey>clark-marshall-1978-reference-diaries</bibkey>
+    </paper>
+    <paper id="9">
+      <!-- https://aclanthology.org/T78-1009/ -->
+      <title>Subsequent Reference: Syntactic and Rhetorical Constraints</title>
+      <author id="david-d-mcdonald"><first>David D.</first><last>McDonald</last></author>
+      <pdf hash="2d424559"/>
+      <bibkey>mcdonald-1978-subsequent-reference</bibkey>
+    </paper>
+    <paper id="10">
+      <!-- https://aclanthology.org/T78-1010/ -->
+      <title>Some Psycholinguistic Constraints on the Construction and Interpretation of Definite Descriptions</title>
+      <author><first>Andrew</first><last>Ortony</last></author>
+      <pdf hash="69807727"/>
+      <bibkey>ortony-1978-psycholinguistic-constraints</bibkey>
+    </paper>
+    <paper id="11">
+      <!-- https://aclanthology.org/T78-1011/ -->
+      <title>Bound Variables and Other Anaphors</title>
+      <author id="barbara-h-partee"><first>Barbara H.</first><last>Partee</last></author>
+      <pdf hash="e8e6a510"/>
+      <bibkey>partee-1978-bound-variables</bibkey>
+    </paper>
+    <paper id="12">
+      <!-- https://aclanthology.org/T78-1012/ -->
+      <title>The Use of Focus as a Tool for Disambiguation of Definite Noun Phrases</title>
+      <author id="candace-l-sidner"><first>Candace L.</first><last>Sidner</last></author>
+      <pdf hash="7b96d2b7"/>
+      <bibkey>sidner-1978-use-focus</bibkey>
+    </paper>
+    <paper id="13">
+      <!-- https://aclanthology.org/T78-1013/ -->
+      <title>Focusing in Dialog</title>
+      <author id="barbara-j-grosz"><first>Barbara J.</first><last>Grosz</last></author>
+      <pdf hash="8c8d4c2e"/>
+      <bibkey>grosz-1978-focusing-dialog</bibkey>
+    </paper>
+    <paper id="14">
+      <!-- https://aclanthology.org/T78-1014/ -->
+      <title>Topic Levels</title>
+      <author><first>Joseph E.</first><last>Grimes</last></author>
+      <pdf hash="f485b2ba"/>
+      <bibkey>grimes-1978-topic-levels</bibkey>
+    </paper>
+    <paper id="15">
+      <!-- https://aclanthology.org/T78-1015/ -->
+      <title>Toward a Rational Model of Discourse Comprehension</title>
+      <author><first>Jerry L.</first><last>Morgan</last></author>
+      <pdf hash="5e12a597"/>
+      <bibkey>morgan-1978-toward-rational</bibkey>
+    </paper>
+    <paper id="16">
+      <!-- https://aclanthology.org/T78-1016/ -->
+      <title>Assent and Compliance in Children’s Language Comprehension</title>
+      <author><first>David R.</first><last>Olson</last></author>
+      <pdf hash="708bc4a0"/>
+      <bibkey>olson-1978-assent-compliance</bibkey>
+    </paper>
+    <paper id="17">
+      <!-- https://aclanthology.org/T78-1017/ -->
+      <title>Speech Acts as a Basis for Understanding Dialogue Coherence</title>
+      <author id="c-raymond-perrault"><first>C. Raymond</first><last>Perrault</last></author>
+      <author id="james-allen"><first>James F.</first><last>Allen</last></author>
+      <pdf hash="d111402b"/>
+      <bibkey>perrault-allen-1978-speech</bibkey>
+    </paper>
+    <paper id="18">
+      <!-- https://aclanthology.org/T78-1018/ -->
+      <title>A Framework for Comparing Language Experiences (with particular emphasis on: The Effect of Audience on Discourse Models)</title>
+      <author><first>Andee</first><last>Rubin</last></author>
+      <pdf hash="5d9a6150"/>
+      <bibkey>rubin-1978-framework-comparing</bibkey>
+    </paper>
+    <paper id="19">
+      <!-- https://aclanthology.org/T78-1019/ -->
+      <title>Intentlonallty and Human Conversations</title>
+      <author id="jaime-g-carbonell"><first>Jaime G.</first><last>Carbonell Jr</last></author>
+      <pdf hash="c50fc619"/>
+      <bibkey>carbonell-jr-1978-intentlonallty-human</bibkey>
+    </paper>
+    <paper id="20">
+      <!-- https://aclanthology.org/T78-1020/ -->
+      <title>On the Interdependence of Language and Perception</title>
+      <author id="david-l-waltz"><first>David L.</first><last>Waltz</last></author>
+      <pdf hash="3c708191"/>
+      <bibkey>waltz-1978-interdependence-language</bibkey>
+    </paper>
+    <paper id="21">
+      <!-- https://aclanthology.org/T78-1021/ -->
+      <title>The Problem of Naming Shapes: Vision-Language Interface</title>
+      <author id="ruzena-bajcsy"><first>R.</first><last>Bajcsy</last></author>
+      <author id="aravind-joshi"><first>A. K.</first><last>Joshi</last></author>
+      <pdf hash="50331b4e"/>
+      <bibkey>bajcsy-joshi-1978-problem-naming</bibkey>
+    </paper>
+    <paper id="22">
+      <!-- https://aclanthology.org/T78-1022/ -->
+      <title>An Argument about the Composition of Conceptual Structure</title>
+      <author><first>Ray</first><last>Jackendoff</last></author>
+      <pdf hash="57d001f8"/>
+      <bibkey>jackendoff-1978-argument-composition</bibkey>
+    </paper>
+    <paper id="23">
+      <!-- https://aclanthology.org/T78-1023/ -->
+      <title>On the Ontological Status of Visual Mental Images</title>
+      <author><first>Stephen Michael</first><last>Kosslyn</last></author>
+      <pdf hash="fd06e265"/>
+      <bibkey>kosslyn-1978-ontological-status</bibkey>
+    </paper>
+    <paper id="24">
+      <!-- https://aclanthology.org/T78-1024/ -->
+      <title>What has language to do with perception? Some speculations on the <i>Lingua Mentis</i></title>
+      <author><first>Zenon W.</first><last>Pylyshyn</last></author>
+      <pdf hash="a2f3a044"/>
+      <bibkey>pylyshyn-1978-language-perception</bibkey>
+    </paper>
+    <paper id="25">
+      <!-- https://aclanthology.org/T78-1025/ -->
+      <title>Semantic Primitives in Language and Vision</title>
+      <author id="yorick-wilks"><first>Yorick</first><last>Wilks</last></author>
+      <pdf hash="b4e13896"/>
+      <bibkey>wilks-1978-semantic-primitives</bibkey>
+    </paper>
+    <paper id="26">
+      <!-- https://aclanthology.org/T78-1026/ -->
+      <title>A Note on Partial Match of Descriptions. Can One Simultaneously Question (Retrieve) and Inform (Update)?</title>
+      <author id="aravind-joshi"><first>Aravind K.</first><last>Joshi</last></author>
+      <pdf hash="74969243"/>
+      <bibkey>joshi-1978-note-partial</bibkey>
+    </paper>
+    <paper id="27">
+      <!-- https://aclanthology.org/T78-1027/ -->
+      <title>With a Spoon in Hand This Must Be the Eating Frame</title>
+      <author><first>Eugene</first><last>Charniak</last></author>
+      <pdf hash="f1be68bb"/>
+      <bibkey>charniak-1978-spoon-hand</bibkey>
+    </paper>
+    <paper id="28">
+      <!-- https://aclanthology.org/T78-1028/ -->
+      <title>Fragments of a Theory of Human Plausible Reasoning</title>
+      <author><first>Allan</first><last>Collins</last></author>
+      <pdf hash="67b4bca2"/>
+      <bibkey>collins-1978-fragments-theory</bibkey>
+    </paper>
+    <paper id="29">
+      <!-- https://aclanthology.org/T78-1029/ -->
+      <title>Indirect Responses to Loaded Questions</title>
+      <author><first>S. Jerrold</first><last>Kaplan</last></author>
+      <pdf hash="135037f9"/>
+      <bibkey>kaplan-1978-indirect-responses</bibkey>
+    </paper>
+    <paper id="30">
+      <!-- https://aclanthology.org/T78-1030/ -->
+      <title>On Reasoning by Default</title>
+      <author><first>Raymond</first><last>Reiter</last></author>
+      <pdf hash="0de53d33"/>
+      <bibkey>reiter-1978-reasoning-default</bibkey>
+    </paper>
+    <paper id="31">
+      <!-- https://aclanthology.org/T78-1031/ -->
+      <title>Path-Based and Node-Based Inference in Semantic Networks</title>
+      <author id="stuart-c-shapiro"><first>Stuart C.</first><last>Shapiro</last></author>
+      <pdf hash="2ddc5535"/>
+      <bibkey>shapiro-1978-path-based</bibkey>
+    </paper>
+    <paper id="32">
+      <!-- https://aclanthology.org/T78-1032/ -->
+      <title>The Representation of Derivable Information in Memory: When What Might Have Been Left Unsaid Is Said</title>
+      <author><first>Rand J.</first><last>Spiro</last></author>
+      <author><first>Joseph</first><last>Esposito</last></author>
+      <author><first>Richard J.</first><last>Vondruska</last></author>
+      <pdf hash="7758fb45"/>
+      <bibkey>spiro-etal-1978-representation-derivable</bibkey>
+    </paper>
+    <paper id="33">
+      <!-- https://aclanthology.org/T78-1033/ -->
+      <title>A Heuristic for Paradigms</title>
+      <author><first>Joseph E.</first><last>Grimes</last></author>
+      <pdf hash="53b42c16"/>
+      <bibkey>grimes-1978-heuristic-paradigms</bibkey>
+    </paper>
+    <paper id="34">
+      <!-- https://aclanthology.org/T78-1034/ -->
+      <title>A Computational Account of Some Constraints on Language</title>
+      <author id="mitch-marcus"><first>Mitchell</first><last>Marcus</last></author>
+      <pdf hash="c9724c43"/>
+      <bibkey>marcus-1978-computational-account</bibkey>
+    </paper>
+    <paper id="35">
+      <!-- https://aclanthology.org/T78-1035/ -->
+      <title>Remarks on Processing, Constraints, and the Lexicon</title>
+      <author id="thomas-wasow"><first>Thomas</first><last>Wasow</last></author>
+      <pdf hash="8beed275"/>
+      <bibkey>wasow-1978-remarks-processing</bibkey>
+    </paper>
+    <paper id="36">
+      <!-- https://aclanthology.org/T78-1036/ -->
+      <title>List of questions suggested for consideration in each session</title>
+      <pdf hash="1950b908"/>
+      <bibkey>nn-1978-list-questions</bibkey>
+    </paper>
+  </volume>
+</collection>
@@ -0,0 +1,85 @@
+import hashlib
+import json
+from pathlib import Path
+import tempfile
+import unittest
+from unittest.mock import patch
+
+from paper_atlas.collectors import aamas, aamas2025
+
+
+FIXTURE = Path(__file__).parent/'fixtures/aamas2025'
+
+
+class Aamas2025Tests(unittest.TestCase):
+    def parse(self, html=None):
+        return aamas2025.parse_contents(html or (FIXTURE/'contents.html').read_text(),
+            {'observed_at': '2026-11-09T12:31:50+00:00', 'url': aamas2025.URLS['contents']})
+
+    def test_real_toc_preserves_every_title_author_and_affiliation(self):
+        expected = json.loads((FIXTURE/'expected-authors.json').read_text())
+        self.assertEqual(hashlib.sha256((FIXTURE/'contents.html').read_bytes()).hexdigest(), expected['source_sha256'])
+        rows, excluded = self.parse()
+        self.assertEqual((len(rows), len(excluded)), (476, 3))
+        by_url = {r['paper_url']: r for r in expected['entries']}
+        for row in rows:
+            original = by_url[row['metadata']['paper_url']]
+            self.assertEqual(row['title'], original['title'])
+            self.assertEqual(row['authors'], [a['name'] for a in original['authors']])
+            self.assertEqual(row['metadata']['author_affiliations'], [a['affiliation'] for a in original['authors']])
+            self.assertIsNone(row['abstract'])
+            self.assertEqual(row['identifiers'], {})
+            self.assertEqual(row['status'], 'accepted')
+            self.assertNotIn('decision_date', row)
+            self.assertTrue(all(t['namespace'] == 'aamas_proceedings:publication_format' for t in row['tags']))
+
+    def test_publication_format_and_track_are_separate(self):
+        rows, excluded = self.parse()
+        groups = {k: sum(r['track'] == k for r in rows) for k in {r['track'] for r in rows}}
+        self.assertEqual(groups, {'proceedings_unspecified': 409, 'blue_sky': 12, 'doctoral': 25, 'demo': 21, 'jaamas': 9})
+        competition = next(r for r in rows if r['external_id'] == '2025/p3000')
+        self.assertEqual(competition['year'], 2025)
+        self.assertEqual(competition['metadata']['publication_format'], 'competition_report')
+        self.assertEqual(competition['metadata']['section_heading'], 'Demo Track')
+        self.assertEqual(competition['track'], 'proceedings_unspecified')
+        self.assertEqual(len({r['external_id'] for r in rows}), 476)
+        self.assertEqual([r['publisher_page'] for r in excluded], [1, 3, 4])
+
+    def test_refuses_changed_edition_inventory_or_author_markup(self):
+        original = (FIXTURE/'contents.html').read_text()
+        for html in [original.replace('AAMAS 2025:', 'AAMAS 2026:', 1),
+                     original.replace('../pdfs/p5.pdf', '../pdfs/p14.pdf', 1),
+                     original.replace('Paper Track</font>', 'Workshop Track</font>'),
+                     original.replace('Amir Abolfazli', '', 1),
+                     original.replace('[COMP24]', '[CHANGED]', 1)]:
+            with self.subTest(change=html[:60]):
+                with self.assertRaises(ValueError): self.parse(html)
+
+    def test_frozen_collection_is_partial_and_never_fetches_pdfs(self):
+        with tempfile.TemporaryDirectory() as tmp, patch.object(aamas2025, 'fetch', side_effect=AssertionError('No HTTP in frozen mode')):
+            out = Path(tmp)/'out'
+            report = aamas.collect(2025, out, snapshots=FIXTURE)
+            self.assertEqual(report['records'], 476)
+            self.assertFalse(report['database_written'])
+            self.assertFalse(report['pdfs_downloaded'])
+            self.assertEqual(len(report['files']), 5)
+            for m in report['files']:
+                self.assertEqual(m['status'], 'partial')
+                self.assertEqual(m['missing_abstract_count'], m['count'])
+                self.assertEqual(hashlib.sha256(Path(m['file']).read_bytes()).hexdigest(), m['file_sha256'])
+            receipt = report['source_receipts'][0]['response_receipt']
+            self.assertEqual(receipt['at_precision'], 'filesystem_write_time_recovered')
+            self.assertTrue((out/'aamas-proceedings-2025.json').is_file())
+
+    def test_rejects_tampered_or_unsafe_frozen_response(self):
+        import shutil
+        for mode in ['tamper', 'path', 'status']:
+            with self.subTest(mode=mode), tempfile.TemporaryDirectory() as tmp:
+                root = Path(tmp); source = root/'source'; shutil.copytree(FIXTURE, source)
+                p = source/'receipts.json'; r = json.loads(p.read_text())
+                if mode == 'tamper': (source/'contents.html').write_text('bad')
+                elif mode == 'path': r['receipts'][0]['file'] = '../index.html'
+                else: r['receipts'][0]['http_status'] = 403
+                p.write_text(json.dumps(r))
+                with self.assertRaises(ValueError): aamas.collect(2025, root/'out', snapshots=source)
+                self.assertFalse((root/'out/aamas-proceedings-2025.json').exists())
@@ -99,7 +99,7 @@ class AamasProceedingsTests(unittest.TestCase):
                aamas.collect(2012, root/'out', snapshots=source)
            self.assertFalse((root/'out/aamas-proceedings-2012.json').exists())
            with self.assertRaises(ValueError):
-                aamas.collect(2025, root/'future', snapshots=source)
+                aamas.collect(2026, root/'future', snapshots=source)
            self.assertFalse((root/'future').exists())

    def test_ids_colliding_across_sections_do_not_produce_a_completed_collection(self):
@@ -0,0 +1,81 @@
+"""Exact AMTCL volume provenance and native metadata/history preservation."""
+import copy
+import hashlib
+import json
+from pathlib import Path
+import sqlite3
+import tempfile
+import unittest
+import xml.etree.ElementTree as ET
+
+from paper_atlas.collectors.acl import collect_xml, volume_scope, text
+from paper_atlas.collectors.acl_historical import event_scope
+from paper_atlas.ingestion.records import ingest_file
+
+FIXTURE = Path(__file__).parent/'fixtures/acl-1963-earlymt.xml'
+COMMIT = 'a7be623d01c48c61405dec31c871121cdbed5a01'
+
+
+class Acl1963Tests(unittest.TestCase):
+    def setUp(self):
+        self.volume = ET.parse(FIXTURE).getroot().find('volume')
+        self.meta = self.volume.find('meta')
+
+    def test_official_snapshot_and_exact_scope(self):
+        self.assertEqual(hashlib.sha256(FIXTURE.read_bytes()).hexdigest(), '4a4626d792ac507de0281cb358769f7246f4a0d6b6a9176ac02e61545c1635fc')
+        self.assertEqual(len(self.volume.findall('paper')), 34)
+        self.assertEqual(volume_scope(self.meta, '1', '1963.earlymt'), ('ACL', ['ACL', 'AMTCL'], 'proceedings_unspecified'))
+        scope = event_scope(self.meta, '1', '1963.earlymt')
+        self.assertEqual(scope['original_venue'], 'EarlyMT')
+        self.assertEqual(scope['publication_format'], 'published_abstract')
+        self.assertEqual(scope['event_dates'], ['1963-08-25', '1963-08-26'])
+
+    def test_other_earlymt_collections_and_volumes_stay_excluded(self):
+        for collection, volume in [('1962.earlymt', '1'), ('1971.earlymt', '1'), ('1963.earlymt', '2'), ('1963.other', '1')]:
+            with self.subTest(collection=collection, volume=volume):
+                self.assertIsNone(volume_scope(self.meta, volume, collection))
+        self.assertIsNone(volume_scope(self.meta, '1'))
+
+    def test_scope_metadata_drift_and_duplicates_are_rejected(self):
+        for field in ['venue', 'year', 'booktitle', 'address', 'month']:
+            for duplicate in [False, True]:
+                meta = copy.deepcopy(self.meta)
+                if duplicate:
+                    ET.SubElement(meta, field).text = text(meta.find(field))
+                else:
+                    meta.find(field).text = 'changed'
+                with self.subTest(field=field, duplicate=duplicate), self.assertRaises(ValueError):
+                    volume_scope(meta, '1', '1963.earlymt')
+
+    def test_every_original_contribution_survives_collection(self):
+        with tempfile.TemporaryDirectory() as tmp:
+            manifest, = collect_xml(FIXTURE, tmp, COMMIT)
+            rows = [json.loads(raw) for raw in Path(manifest['file']).read_text().splitlines()]
+        self.assertEqual(manifest['count'], 34)
+        self.assertEqual(manifest['missing_abstract_count'], 0)
+        self.assertEqual(manifest['excluded_materials'], [])
+        for paper, row in zip(self.volume.findall('paper'), rows):
+            self.assertEqual(row['external_id'], '1963.earlymt-1.'+paper.attrib['id'])
+            self.assertEqual(row['title'], text(paper.find('title')))
+            self.assertEqual(row['abstract'], text(paper.find('abstract')))
+            self.assertEqual(row['authors'], [{'given':text(a.find('first')), 'family':text(a.find('last'))} for a in paper.findall('author')])
+            self.assertEqual(row['metadata']['venue_codes'], ['earlymt'])
+            self.assertEqual(row['metadata']['publication_format'], 'published_abstract')
+            self.assertEqual(row['tags'], [{'namespace':'acl_anthology:publication_format','value':'published_abstract'}])
+            self.assertEqual((row['venue'], row['track'], row['status']), ('ACL', 'proceedings_unspecified', 'accepted'))
+
+    def test_native_ingestion_is_idempotent_and_preserves_abstracts_and_history(self):
+        with tempfile.TemporaryDirectory() as tmp:
+            root = Path(tmp)
+            manifest, = collect_xml(FIXTURE, root/'source', COMMIT)
+            db = root/'sample.sqlite'
+            ingest_file(db, manifest['file'])
+            ingest_file(db, manifest['file'])
+            with sqlite3.connect(db) as c:
+                for table in ['papers', 'source_records', 'submissions', 'decisions']:
+                    self.assertEqual(c.execute('SELECT count(*) FROM '+table).fetchone()[0], 34)
+                self.assertEqual(c.execute("SELECT count(*) FROM papers WHERE abstract IS NOT NULL AND abstract!=''").fetchone()[0], 34)
+                self.assertEqual(c.execute("SELECT DISTINCT status FROM decisions").fetchall(), [('accepted',)])
+                self.assertEqual(c.execute("SELECT v.acronym,e.year,e.track FROM editions e JOIN venues v ON v.id=e.venue_id").fetchall(), [('ACL',1963,'proceedings_unspecified')])
+                self.assertIsNone(c.execute("SELECT 1 FROM venue_aliases WHERE alias='EarlyMT'").fetchone())
+                self.assertEqual(c.execute("SELECT count(*) FROM paper_tags pt JOIN tags t ON t.id=pt.tag_id WHERE t.namespace='acl_anthology:publication_format' AND t.value='published_abstract'").fetchone()[0], 34)
@@ -0,0 +1,114 @@
+"""Pinned T78 evidence, edition-local aliases and an explicitly identified material."""
+import copy
+import hashlib
+import json
+from pathlib import Path
+import sqlite3
+import tempfile
+import unittest
+import xml.etree.ElementTree as ET
+
+from paper_atlas.collectors.acl import collect_xml, volume_scope
+from paper_atlas.collectors.acl_historical import event_scope, material_reason
+from paper_atlas.db.store import Store
+from paper_atlas.ingestion.records import ingest_file
+
+FIXTURE = Path(__file__).parent/'fixtures/acl-t78.xml'
+COMMIT = 'a7be623d01c48c61405dec31c871121cdbed5a01'
+
+
+class AclHistoricalTests(unittest.TestCase):
+    def setUp(self):
+        self.root = ET.parse(FIXTURE).getroot()
+        self.volume = self.root.find('volume')
+        self.meta = self.volume.find('meta')
+
+    def test_frozen_volume_matches_official_snapshot(self):
+        self.assertEqual(hashlib.sha256(FIXTURE.read_bytes()).hexdigest(), 'f712f94cff0cba05f786b7ac31dad2ea9408fde2a17a04531130df623f57dd84')
+        self.assertEqual(len(self.volume.findall('paper')), 36)
+        self.assertEqual(volume_scope(self.meta, '1', 'T78'), ('ACL+TINLAP', ['ACL', 'TINLAP'], 'proceedings_unspecified'))
+
+    def test_unknown_year_collection_or_volume_is_not_globally_aliased(self):
+        for collection, volume in [('T75', '1'), ('T87', '1'), ('T78', '2'), ('C65', '1')]:
+            with self.subTest(collection=collection, volume=volume):
+                self.assertIsNone(event_scope(self.meta, volume, collection))
+                self.assertIsNone(volume_scope(self.meta, volume, collection))
+        self.assertIsNone(volume_scope(self.meta, '1'))
+
+    def test_known_volume_scope_rejects_critical_metadata_drift(self):
+        for field, replacement in [('venue', 'acl'), ('year', '1979'), ('booktitle', 'Another TINLAP'), ('editor/first', 'Another')]:
+            changed = copy.deepcopy(self.meta)
+            changed.find(field).text = replacement
+            with self.subTest(field=field), self.assertRaises(ValueError):
+                event_scope(changed, '1', 'T78')
+        changed = copy.deepcopy(self.meta)
+        ET.SubElement(changed, 'year').text = '1978'
+        with self.assertRaises(ValueError):
+            event_scope(changed, '1', 'T78')
+
+    def test_evidence_retains_original_series_dates_and_role_boundary(self):
+        scope = event_scope(self.meta, '1', 'T78')
+        self.assertEqual(scope['original_venue'], 'TINLAP')
+        self.assertEqual(scope['event_dates'], ['1978-07-25', '1978-07-27'])
+        self.assertEqual([e['pdf_pages'] for e in scope['evidence']], [[9, 10], [3]])
+        self.assertIn('not identified', scope['track_scope_note'])
+        self.assertIn('pending', scope['evidence'][0]['scope'])
+
+    def test_only_exact_session_questions_material_is_excluded(self):
+        paper = copy.deepcopy(self.volume.findall('paper')[-1])
+        self.assertIn('Session question list', material_reason('T78', '1', paper))
+        self.assertIsNone(material_reason('C65', '1', paper))
+        self.assertIsNone(material_reason('T78', '2', paper))
+        paper.set('id', '37')
+        self.assertIsNone(material_reason('T78', '1', paper))
+        paper.set('id', '36')
+        ET.SubElement(paper, 'author')
+        with self.assertRaises(ValueError):
+            material_reason('T78', '1', paper)
+
+    def test_all_authored_source_fields_and_material_receipt_survive_collection(self):
+        with tempfile.TemporaryDirectory() as temporary:
+            manifest, = collect_xml(FIXTURE, temporary, COMMIT)
+            rows = [json.loads(raw) for raw in Path(manifest['file']).read_text().splitlines()]
+        self.assertEqual(manifest['count'], 35)
+        self.assertEqual(manifest['xml_paper_elements'], 36)
+        self.assertEqual(manifest['missing_abstract_count'], 35)
+        self.assertEqual(manifest['excluded_materials'][0]['external_id'], 'T78-1036')
+        for number, row in enumerate(rows, 1):
+            paper = self.volume.findall('paper')[number-1]
+            self.assertEqual(row['external_id'], f'T78-1{number:03d}')
+            self.assertEqual(row['title'], ' '.join(''.join(paper.find('title').itertext()).split()))
+            self.assertEqual(row['metadata']['venue_codes'], ['tinlap'])
+            self.assertEqual(row['metadata']['historical_event_scope']['original_venue'], 'TINLAP')
+            self.assertEqual(row['metadata']['metadata_commit'], COMMIT)
+            self.assertEqual((row['status'], row['track'], row['abstract']), ('accepted', 'proceedings_unspecified', None))
+
+    def test_old_coling_scope_is_unchanged_by_calendar_replacement_evidence(self):
+        meta = ET.fromstring('<meta><venue>coling</venue><year>1965</year><booktitle>COLING 1965</booktitle></meta>')
+        self.assertEqual(volume_scope(meta, '1', 'C65'), ('COLING', ['COLING'], 'main'))
+        self.assertIsNone(event_scope(meta, '1', 'C65'))
+
+    def test_native_ingestion_does_not_alias_other_tinlap_years_or_infer_negative_status(self):
+        with tempfile.TemporaryDirectory() as temporary:
+            root = Path(temporary)
+            manifest, = collect_xml(FIXTURE, root/'source', COMMIT)
+            db = root/'sample.sqlite'
+            store = Store(db)
+            pid, rid = store.record('test_control', 'tinlap1975', 'https://example.org/tinlap1975',
+                                    'An earlier TINLAP contribution', None, ['Jane Doe'])
+            store.decision(pid, rid, 'TINLAP', 1975, 'main', 'tinlap1975', 'https://example.org/tinlap1975', 'accepted', 'test control')
+            store.conn.commit()
+            store.conn.close()
+            ingest_file(db, manifest['file'])
+            ingest_file(db, manifest['file'])
+            with sqlite3.connect(db) as c:
+                self.assertEqual(c.execute('SELECT count(*) FROM papers').fetchone()[0], 36)
+                self.assertEqual(c.execute('SELECT count(*) FROM submissions').fetchone()[0], 36)
+                self.assertEqual(c.execute('SELECT count(*) FROM decisions').fetchone()[0], 36)
+                self.assertEqual(c.execute("SELECT count(*) FROM decisions WHERE status!='accepted'").fetchone()[0], 0)
+                self.assertIsNone(c.execute("SELECT 1 FROM source_records WHERE external_id='T78-1036'").fetchone())
+                aliases = c.execute('SELECT v.acronym,a.alias FROM venue_aliases a JOIN venues v ON v.id=a.venue_id ORDER BY a.alias').fetchall()
+                self.assertEqual(aliases, [('ACL+TINLAP', 'ACL'), ('ACL+TINLAP', 'TINLAP')])
+                metadata = json.loads(c.execute("SELECT metadata_json FROM source_records WHERE external_id='T78-1001'").fetchone()[0])
+                self.assertEqual(metadata['historical_event_scope']['event_year'], 1978)
+                self.assertEqual(c.execute('PRAGMA foreign_key_check').fetchall(), [])
@@ -0,0 +1,148 @@
+"""Historical accepted lists, repeated presentations and preserved work histories."""
+import hashlib
+import json
+from pathlib import Path
+import sqlite3
+import tempfile
+import unittest
+from unittest.mock import patch
+
+from paper_atlas.collectors import iclr_archive as archive
+from paper_atlas.db.store import Store
+from paper_atlas.ingestion.records import ingest_file
+
+OBSERVED = '2026-11-09T00:00:00Z'
+
+
+def entry(identifier, title='A shared research paper', display='Jane Doe and Bob Smith'):
+    return f'<li><div><a href="http://arxiv.org/abs/{identifier}">{title}</a>, {display}</div></li>'
+
+
+def page2015():
+    return ('<h3>Main Conference - Oral Presentations</h3><ol>'+entry('1412.0001v2')+'</ol>'
+            '<h3>Main Conference - Poster Presentations</h3><ol>'+entry('1412.0001v2')+'</ol>'
+            '<h3>Workshop Papers</h3><ol>'+entry('1412.0002', 'A workshop paper')+'</ol>')
+
+
+class IclrArchiveTests(unittest.TestCase):
+    def frozen(self, root, year, html):
+        root.mkdir()
+        data = html.encode()
+        (root/'page.html').write_bytes(data)
+        receipt = {'url': archive.PAGES[year][0], 'file': 'page.html', 'status': 200,
+                   'bytes': len(data), 'sha256': hashlib.sha256(data).hexdigest(), 'finished_at': OBSERVED}
+        (root/'receipts.json').write_text(json.dumps({'receipts': [receipt]}))
+        return root
+
+    def test_oral_poster_repeat_is_one_attempt_with_all_source_variants(self):
+        rows = archive.parse_page(page2015(), 2015, {'observed_at': OBSERVED})
+        main, workshop = rows
+        self.assertEqual((main['track'], workshop['track']), ('main', 'workshop'))
+        self.assertEqual(main['identifiers'], {'arxiv': '1412.0001'})
+        self.assertEqual(main['external_id'], '2015/main/1412.0001')
+        self.assertEqual(main['metadata']['presentations'], ['oral', 'poster'])
+        self.assertEqual(len(main['metadata']['source_variants']), 2)
+        self.assertEqual(main['metadata']['source_variants'][0]['version'], 'v2')
+        self.assertEqual(main['authors'], ['Jane Doe', 'Bob Smith'])
+        self.assertTrue(all(r['status'] == 'accepted' and r['abstract'] is None for r in rows))
+
+    def test_2014_author_span_outside_paragraph_and_trailing_separator(self):
+        html = ('<h3>Conference Proceedings</h3><p>conference papers accepted</p>'
+                '<div><b><a href="http://arxiv.org/abs/1312.0001">A <i>research</i> title</a></b></div>'
+                '<p><span style="font-style: italic">Jane Doe; </span>'
+                '<span style="font-style:italic">Bob Smith; Carol Wu;</span></p>'
+                '<div><a href="http://arxiv.org/abs/1312.0002">Another title</a>'
+                '<div><i><span>First Author; </span><span>Second Author</span></i></div></div>')
+        row, unwrapped = archive.parse_page(html, 2014, {'observed_at': OBSERVED})
+        self.assertEqual(row['title'], 'A research title')
+        self.assertEqual(row['authors'], ['Jane Doe', 'Bob Smith', 'Carol Wu'])
+        self.assertEqual(row['metadata']['source_variants'][0]['authors_display'], 'Jane Doe; Bob Smith; Carol Wu;')
+        self.assertEqual(row['track'], 'main')
+        self.assertEqual(unwrapped['authors'], ['First Author', 'Second Author'])
+
+    def test_same_arxiv_in_different_tracks_keeps_separate_source_attempts(self):
+        html = page2015().replace('1412.0002', '1412.0001v3')
+        rows = archive.parse_page(html, 2015, {'observed_at': OBSERVED})
+        self.assertEqual(len(rows), 2)
+        self.assertEqual({r['external_id'] for r in rows}, {'2015/main/1412.0001', '2015/workshop/1412.0001'})
+        self.assertEqual(rows[1]['metadata']['source_variants'][0]['version'], 'v3')
+
+    def test_2016_code_links_and_breaks_do_not_become_authors(self):
+        html = ('<h1>Accepted Papers (Conference Track)</h1><h3>Oral Presentations</h3><ol>'
+                '<li><a href="http://arxiv.org/abs/1511.00001">A title</a>'
+                '<a href="https://example.org/code">[code]</a><br>Jane Doe, Bob Smith</li></ol>'
+                '<h3>Poster Presentations</h3>')
+        row, = archive.parse_page(html, 2016, {'observed_at': OBSERVED})
+        self.assertEqual(row['authors'], ['Jane Doe', 'Bob Smith'])
+        self.assertEqual(row['metadata']['presentations'], ['oral'])
+
+    def test_conflicting_repeat_unknown_sections_and_unparsed_links_fail(self):
+        variants = [page2015().replace('Poster Presentations', 'Other Presentations'),
+                    page2015().replace('</ol>', '</ol><a href="http://arxiv.org/abs/1412.0003">stray</a>', 1),
+                    page2015().replace('Jane Doe and Bob Smith', 'Different Author', 1)]
+        for html in variants:
+            with self.subTest(html=html), self.assertRaises(ValueError):
+                archive.parse_page(html, 2015, {'observed_at': OBSERVED})
+
+    def test_arxiv_versions_are_preserved_and_unsafe_links_rejected(self):
+        self.assertEqual(archive.arxiv_link('https://arxiv.org/pdf/1511.00001v3.pdf'), ('1511.00001', 'v3'))
+        self.assertIsNone(archive.arxiv_link('https://arxiv.org.example.org/abs/1511.00001'))
+        for url in ['https://arxiv.org/abs/1511.00001?other=1', 'ftp://arxiv.org/abs/1511.00001',
+                    'https://user@arxiv.org/abs/1511.00001', 'https://arxiv.org/abs/not-a-paper']:
+            with self.subTest(url=url), self.assertRaises(ValueError):
+                archive.arxiv_link(url)
+
+    def test_receipt_hash_status_path_and_symlink_cannot_escape_snapshot(self):
+        for failure in ['hash', 'status', 'path', 'symlink']:
+            with self.subTest(failure=failure), tempfile.TemporaryDirectory() as temporary:
+                root = Path(temporary)
+                source = self.frozen(root/'source', 2015, page2015())
+                file = source/'receipts.json'
+                data = json.loads(file.read_text())
+                receipt = data['receipts'][0]
+                if failure == 'hash': receipt['sha256'] = '0'*64
+                elif failure == 'status': receipt['status'] = 404
+                elif failure == 'path': receipt['file'] = '../outside.html'
+                else:
+                    (root/'outside.html').write_text(page2015())
+                    (source/'page.html').unlink()
+                    (source/'page.html').symlink_to(root/'outside.html')
+                file.write_text(json.dumps(data))
+                with self.assertRaises(ValueError):
+                    archive.read_page(2015, snapshots=source)
+
+    def test_changed_inventory_leaves_no_completed_output(self):
+        with tempfile.TemporaryDirectory() as temporary:
+            root = Path(temporary)
+            source = self.frozen(root/'source', 2015, page2015())
+            with self.assertRaises(ValueError):
+                archive.collect(2015, root/'out', snapshots=source)
+            self.assertFalse((root/'out/iclr-archive-2015.json').exists())
+
+    def test_native_import_keeps_rejected_to_accepted_history_and_abstract(self):
+        with tempfile.TemporaryDirectory() as temporary:
+            root = Path(temporary)
+            source = self.frozen(root/'source', 2015, page2015())
+            with patch.dict(archive.PAGES, {2015: (archive.PAGES[2015][0], {'main': 1, 'workshop': 1})}):
+                result = archive.collect(2015, root/'out', snapshots=source)
+                with self.assertRaises(FileExistsError):
+                    archive.collect(2015, root/'out', snapshots=source)
+            db = root/'sample.sqlite'
+            store = Store(db)
+            pid, rid = store.record('arxiv', '1412.0001', 'https://arxiv.org/abs/1412.0001',
+                                    'A shared research paper', 'Keep this canonical abstract.',
+                                    ['Jane Doe', 'Bob Smith'], {'arxiv': '1412.0001'}, {})
+            store.decision(pid, rid, 'Earlier Conference', 2014, 'main', 'original-attempt',
+                           'https://example.org/rejected', 'rejected', 'explicit earlier decision', observed_at=OBSERVED)
+            store.conn.commit()
+            store.conn.close()
+            for manifest in result['files']:
+                ingest_file(db, manifest['file'])
+                ingest_file(db, manifest['file'])
+            with sqlite3.connect(db) as c:
+                self.assertEqual(c.execute('SELECT count(*) FROM papers').fetchone()[0], 2)
+                self.assertEqual(c.execute('SELECT abstract FROM papers WHERE id=?', (pid,)).fetchone()[0], 'Keep this canonical abstract.')
+                history = c.execute('SELECT v.acronym,e.year,d.status FROM decisions d JOIN submissions s ON s.id=d.submission_id JOIN editions e ON e.id=s.edition_id JOIN venues v ON v.id=e.venue_id WHERE s.paper_id=? ORDER BY e.year', (pid,)).fetchall()
+                self.assertEqual(history, [('Earlier Conference', 2014, 'rejected'), ('ICLR', 2015, 'accepted')])
+                self.assertEqual(c.execute('SELECT count(*) FROM submissions').fetchone()[0], 3)
+                self.assertEqual(c.execute('PRAGMA foreign_key_check').fetchall(), [])
Author	SHA1	Message	Date
codex-bot	8b1e983e7d	Recognize the exact 1963 AMTCL annual meeting volume Metadata integrity checks / metadata (push) Successful in 2m10s Details	2026-11-09 23:21:53 +08:00
Zayden-Jung	b226b3b98b	Merge pull request '限定ACL1978与TINLAP卷的来源关系及非论文条目' (#42 ) from codex/acl-early into main Metadata integrity checks / metadata (push) Successful in 2m18s Details Reviewed-on: #42	2026-11-09 13:42:18 +00:00
Zayden-Jung	05ada8daf8	Merge pull request '支持 AAMAS 2025 官方出版目录元数据采集' (#44 ) from codex/aamas2025 into main Metadata integrity checks / metadata (push) Successful in 3m20s Details Reviewed-on: #44	2026-11-09 13:42:00 +00:00
codex-bot	13408869c2	Add reviewed AAMAS 2025 publisher metadata adapter Metadata integrity checks / metadata (push) Successful in 2m11s Details	2026-11-09 20:50:29 +08:00
codex-bot	443a79dce2	Preserve edition-specific ACL and TINLAP 1978 source evidence Metadata integrity checks / metadata (push) Successful in 2m6s Details	2026-11-09 17:20:27 +08:00
Zayden-Jung	ec9edd88d4	Merge pull request '保留ICLR历史录用档案的轨道与完整来源证据' (#41 ) from codex/iclr-archive into main Metadata integrity checks / metadata (push) Successful in 2m54s Details Reviewed-on: #41	2026-11-09 09:02:49 +00:00
codex-bot	b5c0618be5	Handle standalone italic author blocks in the historical archive Metadata integrity checks / metadata (push) Successful in 2m6s Details	2026-11-09 16:24:42 +08:00
codex-bot	81864e48ae	Preserve complete fragmented ICLR 2014 author paragraphs Metadata integrity checks / metadata (push) Successful in 2m9s Details	2026-11-09 16:19:19 +08:00
Zayden-Jung	8ad052f577	Merge pull request '核验学习与规划会议前身卷并补齐发现范围' (#40 ) from codex/historical-predecessors into main Metadata integrity checks / metadata (push) Successful in 2m12s Details Reviewed-on: #40	2026-11-09 08:13:52 +00:00
codex-bot	348ebe0695	Collect archived ICLR accepted lists with track and presentation evidence Metadata integrity checks / metadata (push) Successful in 2m15s Details	2026-11-09 16:09:13 +08:00