Diminishing Returns in Verification Accuracy from Scaling Diverse Debating Agents on the FEVER-LC Benchmark

Large Language Models (LLMs) suffer from hallucinations and factual inaccuracies, especially in complex reasoning and fact verification tasks. Multi-Agent Debate (MAD) systems aim to improve answer accuracy by enabling multiple LLM agents to engage in dialogue, promoting diverse reasoning and mutual verification. However, existing MAD frameworks primarily rely on internal knowledge or static documents, making them vulnerable to hallucinations. While MADKE introduces external evidence to mitigate this, its one-time retrieval mechanism limits adaptability to new arguments or emerging information

Research goal: Does scaling the number of debating agents with diverse retrieval strategies yield diminishing returns in verification accuracy on the FEVER-LC benchmark?

Autonomous synthesis report generated by Assignee Research. Tribunal consensus score: 7.8/10.