大模型
arxiv.org ·
探测大型语言模型中的知识归因
论文研究“贡献性归因”，即判断 LLM 每次输出主要由提供的上下文还是内部知识驱动，以区分忠实性违规与事实性违规两类幻觉。作者提出 AttriWiki 自监督流程，通过让模型从记忆回忆被隐去的实体或从上下文读取来生成标注数据，无需依赖知识冲突。在 Llama-3.1-8B、Mistral-7B、Qwen-7B 上训练的线性探针最高达 0.96 Macro-F1，迁移到 SQuAD 与 WebQuestions 为 0.94-0.99 Macro-F1，并零样本泛化到 Tighidet 等（2024）的基准，在冲突设置下无需重训即优于其探针。归因不匹配会使错误率上升最多 70%，但正确归因并不保证答案正确。
AI推荐理由 arXiv 论文（v3，2026-09-10 修订），提出 AttriWiki 自监督流程生成标注数据，用线性探针区分 LLM 输出由上下文还是内部知识驱动；在 Llama-3.1-8B、Mistral-7B、Qwen-7B 上达 0.96 Macro-F1，迁移到 SQuAD/WebQuestions 为 0.94-0.99，并零样本泛化到 Tighidet 等（2024）基准。局限：归因正确不保证答案正确，作者指出仍需更广的检测框架；摘要未提供代码或数据链接。
78
正文
Computer Science > Computation and Language
arXiv:2602.22787 (cs)
[Submitted on 26 Feb 2026 (v1), last revised 10 Sep 2026 (this version, v3)]
Title:Probing for Knowledge Attribution in Large Language Models
Authors:Ivo Brink, Alexander Boer, Dennis Ulmer
View a PDF of the paper titled Probing for Knowledge Attribution in Large Language Models, by Ivo Brink and 2 other authors
View PDF
HTML (experimental)
Abstract:Large language model (LLM) hallucinations, meaning fluent but factually incorrect generations, fall into two types: faithfulness violations, where the model misuses provided context, and factuality violations, where answers reflect errors in internal knowledge. Proper mitigation depends on knowing which source drives each answer. We study contributive attribution, i.e. the classification of the dominant knowledge source behind each output, and show that a simple linear probe trained on hidden representations can reliably identify it. We introduce AttriWiki, a self-supervised pipeline that automatically generates labelled training data by prompting models to recall withheld entities from memory or read them from context without relying on knowledge conflicts. Probes trained on AttriWiki achieve up to 0.96 Macro-$F_1$ on Llama-3.1-8B, Mistral-7B, and Qwen-7B, transfer to SQuAD and WebQuestions with 0.94-0.99 Macro-$F_1$, and generalise zero-shot to Tighidet et al. (2024)'s benchmark, outperforming their probe on conflicting settings without retraining. Furthermore, attribution mismatches raise error rates by up to 70%, though correct attribution does not guarantee correct answers, pointing to the need for broader detection frameworks.
Subjects:
Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
Cite as:
arXiv:2602.22787 [cs.CL]
(or
arXiv:2602.22787v3 [cs.CL] for this version)
https://doi.org/10.48550/arXiv.2602.22787
Focus to learn more
arXiv-issued DOI via DataCite
Submission history
From: Dennis Ulmer [view email]
[v1]
Thu, 26 Feb 2026 09:21:12 UTC (20,307 KB)
[v2]
Wed, 27 May 2026 16:19:23 UTC (20,310 KB)
[v3]
Thu, 10 Sep 2026 12:57:27 UTC (20,314 KB)
Full-text links:
Access Paper:
View a PDF of the paper titled Probing for Knowledge Attribution in Large Language Models, by Ivo Brink and 2 other authors
View PDF
HTML (experimental)
TeX Source
view license
Current browse context:
cs.CL
new
|
recent
| 2026-02
Change to browse by:
cs
cs.AI
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?)