arXiv cs.CL· Angana Borah, Zoha Hayat Bhatti, Zhijing Jin, Rada Mihalcea·· 11 小时前AI 评分31
CuriousTogether：多智能体 LLM 交互中的好奇心与文化根基研究
CuriousTogether: Implications of Curiosity and Cultural Grounding in Multi-Agent LLM Interactions
AI 导读
研究团队提出 CuriousTogether 框架，从求新、认知好奇与社会好奇三个维度评估好奇心驱动的多智能体 LLM 交互。在覆盖推理、创意生成和社会偏见缓解的八项任务中，好奇心引导的多智能体设置在七项上优于无好奇心设置，故事生成的语义多样性最高提升 20.6%，刻板印象率下降 14.9 个百分点。
正文 · 原文
Abstract:Large Language Models (LLMs) are increasingly deployed in multi-agent settings, exhibiting emergent collective behaviors, yet the role of curiosity as a driver of such dynamics remains understudied. We introduce CuriousTogether, a framework evaluating curiosity-driven multi-agent LLM interactions across three complementary dimensions: novelty-seeking, epistemic and social curiosity. Across eight tasks spanning reasoning, creative generation, and social bias mitigation, curiosity-guided multi-agent settings yield improvements over non-curious ones on seven out of eight tasks, with gains of up to 20.6% in story generation (semantic diversity) and a 14.9 pp reduction in stereotype rates. Furthermore, country-conditioned cultural grounding show benefits for story generation and social tasks, but can hurt precision-oriented reasoning. Finally, we show that inference-time multi-agent curiosity-based approach offers favorable quality-token trade-offs, compared with single-agent baselines. Together, these findings support curiosity-guided interaction as a practical mechanism for improving multi-agent collaboration, with downstream benefits for reasoning, creative, and societal tasks.
| Comments: | Preprint under review |
| Subjects: | Computation and Language (cs.CL); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2611.01301 [cs.CL] |
| (or arXiv:2611.01301v1 [cs.CL] for this version) | |
| https://doi.org/10.48550/arXiv.2611.01301 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Angana Borah [view email]
[v1]
Sat, 31 Oct 2026 22:01:33 UTC (403 KB)
来源：arXiv cs.CL · arxiv.org