paperSeptember 2026Unreviewed

Remember and Reweight: Enhancing Multi-Agent Debate with Experience Memory and Confidence Estimation

Xuanfa Jin, Zhijian Ma, Yongcheng Zeng, Xinyu Cui, Haifeng Zhang, Jun Wang

Abstract

Multi-agent debate (MAD) improves the reasoning capabilities of large language models by having multiple agents iteratively refine their responses through discussion. However, MAD suffers from a critical vulnerability known as shared misconception: when a majority of agents initially converge on an incorrect answer, the debate process tends to amplify rather than correct the error. Existing methods primarily address peer skew but leave the agents' inherently biased concept priors unaddressed. To

Cite

@misc{jin2026remember,
  title = {{Remember and Reweight: Enhancing Multi-Agent Debate with Experience Memory and Confidence Estimation}},
  author = {Xuanfa Jin and Zhijian Ma and Yongcheng Zeng and Xinyu Cui and Haifeng Zhang and Jun Wang},
  year = {2026},
  month = sep,
  eprint = {2609.03619},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2609.03619}
}