arXiv cs.CL· Minsu Oh, Seunghui Jwa, Soo Kyun Kim, Yeo-Chan Yoon·· 12 小时前AI 评分31
研究：采集日期元数据会改变时序问答的答案
Collection-Date Assignment Redirects Answers in Temporal QA
AI 导读
在 595 个 Wikipedia 页面上交换问题日与次日的采集日期、保持证据不变，五个开源权重 LLM 的后续引用精确匹配（EM）下降 37.6-51.4 个百分点，并在 128 个不重叠页面上复现。
正文 · 原文
Abstract:A collection date records when a source revision was acquired, not when its contents changed. Can assigning this metadata redirect temporal question answering despite unchanged evidence? Swapping the question day and following day between two revisions lowers later-reference exact match (EM) by 37.6-51.4 percentage points across five open-weight LLMs on 595 Wikipedia pages, replicated on 128 disjoint pages. Previous-day swaps retain 29.0-39.6-point losses on 497 pages. Seven prompt variants, including explicit ignore-collection instructions and archived-revision provenance, retain the effect. Fictional answers and English month-name question dates show that remembered answers and identical date strings are unnecessary. Matched thinking reduces Qwen3.5's penalty from 30.1 to 11.1 points, while native reasoners vary. Two API LLMs show small Wikipedia swap effects and rarely shift to earlier answers on fictional records. Large swap-induced losses concentrate around the question date; swaps between distant dates have much smaller effects. Adding fields containing one observed acquisition date lowers EM relative to omission in three of five LLMs after correction. The diagnostic isolates metadata effects from changing evidence or missing the relevant revision, and provides a practical audit for temporal QA inputs.
| Subjects: | Computation and Language (cs.CL) |
| Cite as: | arXiv:2610.32191 [cs.CL] |
| (or arXiv:2610.32191v1 [cs.CL] for this version) | |
| https://doi.org/10.48550/arXiv.2610.32191 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Minsu Oh [view email]
[v1]
Fri, 23 Oct 2026 05:37:24 UTC (88 KB)
来源：arXiv cs.CL · arxiv.org