Can LLMs Produce Faithful Explanations For Fact-checking? Towards Faithful Explainable Fact-Checking via Multi-Agent Debate
read the original abstract
Fact-checking research has extensively explored verification but less so the generation of natural-language explanations, crucial for user trust. While Large Language Models (LLMs) excel in text generation, their capability for producing faithful explanations in fact-checking remains underexamined. Our study investigates LLMs' ability to generate such explanations, finding that zero-shot prompts often result in unfaithfulness. To address these challenges, we propose the Multi-Agent Debate Refinement (MADR) framework, leveraging multiple LLMs as agents with diverse roles in an iterative refining process aimed at enhancing faithfulness in generated explanations. MADR ensures that the final explanation undergoes rigorous validation, significantly reducing the likelihood of unfaithful elements and aligning closely with the provided evidence. Experimental results demonstrate that MADR significantly improves the faithfulness of LLM-generated explanations to the evidence, advancing the credibility and trustworthiness of these explanations.
This paper has not been read by Pith yet.
Forward citations
Cited by 8 Pith papers
-
RW-Post: Auditable Evidence-Grounded Multimodal Fact-Checking in the Wild
RW-Post is a new post-aligned text-image benchmark with auditable annotations from real fact-checks that reveals current LVLMs struggle with faithful evidence grounding but improve under evidence-bounded evaluation.
-
ChartFI: Benchmarking Faithfulness and Insightfulness of Chart Descriptions from Multimodal Large Language Models
ChartFI-Bench supplies 896 chart-description pairs from visually complex charts and defines four metrics (Faithfulness, Coverage, Informativeness, Acuity) aligned to four quality dimensions to evaluate MLLM-generated ...
-
ChartFI: Benchmarking Faithfulness and Insightfulness of Chart Descriptions from Multimodal Large Language Models
ChartFI-Bench supplies 896 chart-description pairs and four metrics (Faithfulness, Coverage, Informativeness, Acuity) to evaluate MLLM-generated chart descriptions on faithfulness and insightfulness.
-
Whose Story Gets Told? Positionality and Bias in LLM Summaries of Life Narratives
A proposed pipeline shows LLMs introduce detectable race and gender biases when summarizing life narratives, creating potential for representational harm in research.
-
RW-Post: Auditable Evidence-Grounded Multimodal Fact-Checking in the Wild
RW-Post is an auditable benchmark linking social media posts to evidence from human fact-check articles for evaluating multimodal AI fact-checking across different evidence regimes.
-
Debating Truth: Debate-driven Claim Verification with Multiple Large Language Model Agents
DebateCV uses opposing debater agents and a trained moderator agent to verify complex claims via debate, outperforming single-agent baselines in accuracy and justification quality.
-
RW-Post: Auditable Evidence-Grounded Multimodal Fact-Checking in the Wild
RW-Post is an auditable text-image benchmark for real-world multimodal fact-checking that links posts to evidence traces from human fact-check articles and includes the AgentFact baseline for evaluation.
-
Large Language Model Agent: A Survey on Methodology, Applications and Challenges
A survey that deconstructs LLM agent systems via a methodology-centered taxonomy linking design principles to emergent behaviors, applications, and challenges.
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.