Multi-Agent LLMs Fail to Explore Each Other
- 类型:arxiv
- 标识:2607.11250
- 链接:https://arxiv.org/abs/2607.11250
- 主分类:agent
- 形态:method
- 被引:1
- 被引来源:Semantic Scholar
- S2被引:1
- OpenAlex被引:0
- 影响力被引:0
- TLDR:This work introduces Multi- Agent Contextual Exploration (MACE), a lightweight framework that explicitly promotes exploration through structured peer selection that substantially improves exploration behavior and downstream task performance and shows theoretically that the value of exploration increases with agent diversity.
- OpenAlex ID:W7168258747
- OpenAlex DOI:10.48550/arxiv.2607.11250
- DOI:10.48550/arxiv.2607.11250
- DOI来源:OpenAlex
- 开放获取:green
- 开放获取链接:https://doi.org/10.48550/arxiv.2607.11250
- OpenAlex更新:2026-07-19
- 待LLM分类:否
- 标题中文:Multi-Agent LLMs 未能互相探索
- TLDR中文:本文提出 Multi-Agent Contextual Exploration (MACE),一个通过结构化的对等体选择显式促进探索的轻量级框架,显著改善了探索行为和下游任务表现,并在理论上证明探索价值随 Agent 多样性增加而提升。
- 来源文件:
- /inbox/tom/_candidates/2026-07-15-agent-rag-longcontext-candidates.json
- [S2 enrich]
- [OpenAlex backfill]