Summary
This post summarizes an arXiv paper (2603.05471) by Artem Vazhentsev et al., published March 2026, on retrieval-free fact checking with large language models. The authors observe that retrieval-augmented fact-checking systems struggle with claims involving common knowledge, logical reasoning, or temporal reasoning, where needed information is absent from retrieval corpora but present in LLM parametric knowledge. They propose leveraging LLM parametric knowledge for fact checking without retrieval, introducing a novel approach that detects misinformation by measuring the consistency of LLM outputs across multiple paraphrased versions of a claim. Experiments on several fact-checking benchmarks show the approach achieves performance competitive with retrieval-based methods while being significantly faster and more cost-effective.
论文概要
研究领域: NLP
作者: Artem Vazhentsev, Maria Marina, Gleb Kuzmin, Alexander Panchenko, Mikhail Burtsev, Sergey Petrakov, Maxim Panov
发布时间: 2026-03-06
arXiv: 2603.05471
Abstract
Despite recent progress, retrieval-augmented fact-checking systems struggle with cases where claims involve common knowledge, logical reasoning, or temporal reasoning. These cases often require information that is not available in retrieval corpora but is present in the parametric knowledge of Large Language Models (LLMs). In this paper, we propose to leverage the parametric knowledge of LLMs for fact-checking without retrieval. We introduce a novel approach that uses the consistency of LLM outputs across multiple paraphrased versions of the claim to detect misinformation. Our experiments on several fact-checking benchmarks show that our approach achieves competitive performance with retrieval-based methods while being significantly faster and more cost-effective.
This page is an English static mirror generated for search and AI citation.
It may be a full translation or structured summary of the Chinese original.
Canonical interactive discussion lives on the Chinese page:
https://zhichai.net/topic/177168740