发表机构
McGill University; Mila – Québec AI Institute; University of Connecticut; Sichuan University; Gradient; Princeton University（麦吉尔大学; Mila——魁北克人工智能研究所; 康涅狄格大学; 四川大学; Gradient; 普林斯顿大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
研究针对基于大语言模型的智能体在科学协作中存在的问题，提出CoWeaver算法，通过填补能力差距匹配人员，经两阶段排名筛选，结合探索与贪婪策略，在匹配质量和效率上优于基线，能促进人机科学协作。
AI 中文摘要
基于大语言模型的智能体在写作文章、编码和信息检索方面表现出色。然而，由于问题的双向动态性质以及对决策可解释性的高要求，它们在科学界难以形成强大的合作。我们提出了CoWeaver，一种双向、可学习且可解释的算法，用于在人机网络中匹配科学家并形成强大的合作。CoWeaver通过填补能力差距来匹配候选者和请求者，并通过两阶段排名步骤筛选候选者。最后，该模型通过保持不确定性感知能力估计并根据请求者的反馈进行更新来探索新人。我们表明，结合探索（UCB）和贪婪策略的CoWeaver选择机制在20项任务中的6项上超过了仅贪婪策略（理论上的最佳解决方案），在选择最佳候选者方面与仅贪婪策略表现相当。我们在匹配质量和效率方面比较了CoWeaver与基线。CoWeaver在所有指标上均优于基线。
英文摘要
LLM-based agents excel at writing articles, coding and information retrieval. However, they fail to form strong collaborations within the scientific community due to the bidirectional, dynamic nature of the problem and a high demand of decision interpretability. We proposed COWEAVER, a bidirectional, learnable and explainable algorithm to match scientists and form strong collaborations within a human-agent network. COWEAVER matches candidates and requesters through filling capability gaps and filters candidates through a two-stage ranking step. Finally, the model explores newcomers by maintaining uncertainty-aware capability estimates and updating them through requester's feedback. We show that the selection mechanism of combining both exploration (UCB) and greedy of COWEAVER exceeds the greedy-only mechanism - the analytical best solution - on 6 out of the 20 tasks and performed on par with the greedy-only mechanism in terms of selecting the best candidate. We compared COWEAVER baselines in terms of matching quality and efficiency. COWEAVER outperforms baselines on all metrics.
