arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2408.14575cs.AI
收藏

EVINCE:利用条件统计与信息论优化多LLM对话

EVINCE: Optimizing Multi-LLM Dialogues Using Conditional Statistics and Information Theory

Stanford University(斯坦福大学)

Edward Y. Chang

AI总结:

EVINCE利用条件统计和信息论,通过双熵优化动态调节多LLM对话,平衡视角多样性与先验知识,提升协作效率。

摘要:

EVINCE(条件交换中的熵与变异)是一个利用条件统计和信息论优化多LLM对话的新框架。它解决了多智能体辩论(MAS)框架中的局限性,在这些框架中,多个LLM进行聊天而无需行为调节或互信息质量评估。通过使用双熵优化来平衡视角多样性和先验知识,EVINCE提供了定量工具来动态调节LLM的语言行为。当互信息较低且交叉熵和Wasserstein距离均较高时,EVINCE促进对抗性对话以暴露不同视角并揭示不一致性。相反,随着交叉熵降低且互信息趋于稳定,它将讨论过渡到和解阶段,鼓励妥协和承认有效观点。通过使用信息论指标并优化互信息,EVINCE成为多LLM协作的结构化且高效的框架。

英文摘要:

EVINCE (Entropy and Variation IN Conditional Exchanges) is a novel framework for optimizing multi-LLM dialogues using conditional statistics and information theory. It addresses limitations in multi-agent debate (MAS) frameworks, where multiple LLMs chat without behavior modulation or mutual information quality assessment. Using dual entropy optimization to balance perspective diversity and prior knowledge, EVINCE provides quantitative tools to dynamically regulate LLM linguistic behaviors. When mutual information is low and both cross-entropy and Wasserstein distance are high, EVINCE promotes contentious dialogues to expose diverse perspectives and uncover inconsistencies. Conversely, as cross-entropy decreases and mutual information stabilizes, it transitions discussions into a conciliatory phase, encouraging compromise and acknowledgment of valid points. Using information-theoretic metrics and optimizing mutual information, EVINCE emerges as a structured and highly effective framework for multi-LLM collaboration.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑