大模型
arxiv.org ·
寻找一位老朋友：诊断并缓解基于LLM的法条问答中的时间失效模式
该论文研究 LLM 法律问答中的两类时间失效模式：训练截止后的规则过时（模型沿用已被立法修正取代的旧规则）与近因偏差（即使历史版本才适用，模型仍偏好新条款）。作者构建了 312 条经专家验证、时间敏感的德语法条 QA 基准，分为截止后修正问题、修正前问题和多条款修正前问题三类，并在 Vanilla、Web-search 及两种通过事实日期抽取与版本过滤强制时间有效性的 RAG 变体下评估 OpenAI、Anthropic、DeepSeek 的五个模型。结果显示 Vanilla 在截止后场景严重退化，两种 RAG 方法在各题型上均大幅提升，而 web search 收益不稳定且在历史锚定任务上呈现明显近因偏差，表明可靠的法律问答需将时间有效性作为硬约束。
AI推荐理由 arXiv 论文（v2，2026-09-25 修订），提出 312 条经专家验证的德语法条时间敏感 QA 基准，覆盖三类问题，并对比 Vanilla、Web-search 与两种带时间有效性约束的 RAG 设置，评估 OpenAI、Anthropic、DeepSeek 五个模型。具体新信息是：Vanilla 在 post-cutoff 场景严重退化，两种 RAG 均显著改善，而 web search 收益不稳定且对历史锚定任务表现出明显 recency bias。局限在于结论基于德语法条与 LLM-as-a-judge（虽经人工专家校准），未提供跨语言或真实部署验证。
78
正文
Computer Science > Computation and Language
arXiv:2605.23497 (cs)
[Submitted on 22 May 2026 (v1), last revised 25 Sep 2026 (this version, v2)]
Title:Asking For An Old Friend: Diagnosing and Mitigating Temporal Failure Modes in LLM-based Statutory Question Answering
Authors:Max Prior, Andreas Schultz, Matthias Grabmair
View a PDF of the paper titled Asking For An Old Friend: Diagnosing and Mitigating Temporal Failure Modes in LLM-based Statutory Question Answering, by Max Prior and 2 other authors
View PDF
HTML (experimental)
Abstract:Large language models are increasingly used for legal research, yet their fixed training cutoffs and reliance on static parametric knowledge are at odds with the evolving nature of statutory law. We study two temporal failure modes: post-cutoff staleness, where models apply superseded rules after legislative amendments, and recency bias, where models prefer newer provisions even when a historical version governs the fact pattern. To this end, we present a benchmark of 312 expert-validated, time-sensitive German statutory QA pairs spanning three categories: Post-Cutoff Amendment Questions, Pre-Amendment Questions, and Multi-Provision Pre-Amendment Questions. We evaluate five LLMs by OpenAI, Anthropic and DeepSeek under four inference settings: Vanilla, Web-search, and two retrieval-augmented variants that enforce temporal validity via a fact date extraction and version filtering. Using an LLM-as-a-judge validated against human expert ratings, we find severe degradation in the Vanilla post-cutoff setting. Both RAG approaches substantially improve performance across all question types, while web search yields unstable gains and exhibits a marked recency bias on historically anchored tasks. Our results indicate that reliable legal QA requires treating temporal validity as a hard constraint.
Comments:
Subjects:
Computation and Language (cs.CL)
Cite as:
arXiv:2605.23497 [cs.CL]
(or
arXiv:2605.23497v2 [cs.CL] for this version)
https://doi.org/10.48550/arXiv.2605.23497
Focus to learn more
arXiv-issued DOI via DataCite
Submission history
From: Max Prior [view email]
[v1]
Fri, 22 May 2026 11:02:01 UTC (618 KB)
[v2]
Fri, 25 Sep 2026 12:07:11 UTC (618 KB)
Full-text links:
Access Paper:
View a PDF of the paper titled Asking For An Old Friend: Diagnosing and Mitigating Temporal Failure Modes in LLM-based Statutory Question Answering, by Max Prior and 2 other authors
View PDF
HTML (experimental)
TeX Source
view license
Current browse context:
cs.CL
new
|
recent
| 2026-05
Change to browse by:
cs
References & Citations
NASA ADS
Google Scholar
Semantic Scholar
export BibTeX citation
Loading...
BibTeX formatted citation
×
loading...
Data provided by:
Bookmark
Bibliographic Tools
Bibliographic and Citation Tools
Bibliographic Explorer Toggle
Bibliographic Explorer (What is the Explorer?)
Connected Papers Toggle
Connected Papers (What is Connected Papers?)
Litmaps Toggle
Litmaps (What is Litmaps?)
scite.ai Toggle
scite Smart Citations (What are Smart Citations?)
Code, Data, Media
Code, Data and Media Associated with this Article
alphaXiv Toggle
alphaXiv (What is alphaXiv?)
Links to Code Toggle
CatalyzeX Code Finder for Papers (What is CatalyzeX?)
DagsHub Toggle
DagsHub (What is DagsHub?)
GotitPub Toggle
Gotit.pub (What is GotitPub?)
Huggingface Toggle
Hugging Face (What is Huggingface?)
ScienceCast Toggle
ScienceCast (What is ScienceCast?)
Demos
Demos
Replicate Toggle
Replicate (What is Replicate?)
Spaces Toggle
Hugging Face Spaces (What is Spaces?)
Spaces Toggle
TXYZ.AI (What is TXYZ.AI?)
Related Papers
Recommenders and Search Tools
Link to Influence Flower
Influence Flower (What are Influence Flowers?)
Core recommender toggle
CORE Recommender (What is CORE?)
Author
Venue
Institution
Topic
About arXivLabs
arXivLabs: experimental projects with community collaborators
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Which authors of this paper are endorsers? |
Disable MathJax (What is MathJax?)