arXiv NLP
techCenter
Remember and Reweight: Enhancing Multi-Agent Debate with Experience Memory and Confidence Estimationtranslating…
arXiv:2609.03619v1 Announce Type: new
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…
Keywords#Multi-Agent Debate#Confidence#Confidence Estimation#Enhancing#Enhancing Multi-Agent
Comments
Sign in to join the discussion.
Loading comments…