← Back to feed

Verify, Then Vote: Overturning Wrong Majorities in Multi-Agent LLMs

AI · Studies · 2026-11-03 06:16

论文提出“先验证再投票”机制，用于在多智能体大模型协作中识别并推翻错误的多数结论，提升集体决策可靠性。

Source: github.com

Auto-generated summary for reference only. RSS