arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2602.12089cs.GTcs.AIcs.HC
收藏

选择你的代理:在多方谈判中采用AI顾问、教练和代理的权衡

Choose Your Agent: Tradeoffs in Adopting AI Advisors, Coaches, and Delegates in Multi-Party Negotiation

Harvard University(哈佛大学);Google DeepMind Canada(谷歌深Mind加拿大);Google DeepMind United States(谷歌深Mind美国);Google DeepMind New York United States(谷歌深Mind纽约美国)

Kehang Zhu, Nithum Thain, Vivian Tsai, James Wexler, Crystal Qian

AI总结:

研究探讨了在多方谈判中采用AI顾问、教练和代理的权衡,发现尽管参与者更偏好具有更高控制权的顾问,但代理模式能显著提升集体收益,揭示了AI生成提议与人类行为之间的过滤效应。

摘要:

随着AI在社交情境中的使用日益增加,理解代理与用户交互对于设计提升个人和群体结果的系统至关重要。我们进行了一项在线行为实验(N=243),参与者在三人小组中玩三种多轮谈判博弈。每种博弈以随机顺序呈现,提供单一LLM辅助模式:顾问的主动建议、教练的反馈或代理的自主执行。所有三种模式均基于在该谈判环境中超人类表现的LLM。在每一轮中,参与者私密决定是手动操作还是使用该博弈中可用的AI模式。我们记录了偏好与绩效之间的不一致:参与者强烈偏好具有更高控制权的顾问(44%)而非代理(19%),但只有在代理模式下,群体才显著增加集体盈余。调整自愿非合规性后,将任务委托给AI会产生有说服力的个体福利增益,大约是意图治疗估计的1.5倍。机制分析揭示了这一差距源于一个过滤器:AI生成的提议在所有条件下都创造更多的联合盈余,但在线上顾问和教练模式中,用户会修改、覆盖或忽略AI的建议,从而回归人类基线的权衡模式。代理的优势并非源于不同的AI能力,而是绕过了这一过滤步骤。实现这些福利增益不仅取决于模型能力,还取决于该能力如何交付的交互结构。我们主张,辅助模式应被设计为具有内生参与的机制;与自动化辅助改善福利相容的交互规则是必要的前提。

英文摘要:

As AI usage becomes more prevalent in social contexts, understanding agent-user interaction is critical to designing systems that imp rove both individual and group outcomes. We present an online behavioral experiment (N=243) in which participants play three multi-tu rn bargaining games in groups of three. Each game, presented in randomized order, grants access to a single LLM assistance modality: proactive recommendations from an Advisor, reactive feedback from a Coach, or autonomous execution by a Delegate. All three modalitie s are powered by an LLM with super-human performance within this negotiation setting. On each turn, participants privately decide whe ther to act manually or use the AI modality available in that game. We document a preference-performance misalignment: participants s trongly prefer the higher-control Advisor (44%) over the Delegate (19%), yet groups only significantly increase collective surplus un der Delegate access. Adjusting for voluntary non-compliance, delegating to the AI yields suggestive individual welfare gains, roughly 1.5x the intent-to-treat estimate. A mechanism analysis traces this gap to a human filter: AI-generated proposals create more joint surplus than manual proposals across all conditions, but in the Advisor and Coach modes users modify, override, or ignore the AI's su ggestions, reverting toward human-baseline trade patterns. The Delegate advantage arises not from a different AI capability but from bypassing this filtering step altogether. Realizing these welfare gains depends not only on model capability, but on the interaction structure through which that capability is delivered. We argue that assistance modalities should be designed as mechanisms with endog enous participation; adoption-compatible interaction rules are a prerequisite to improving welfare with automated assistance.

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。

↑