用进化算法生成多样真实用户角色，提升大模型代理的实战评估效果。
Beyond Cooperative Simulators: Generating Realistic User Personas for Robust Evaluation of LLM Agents
- 通过进化搜索自动生成多样化用户行为策略，保持任务目标不变。
- 在零售与航空场景中，评估得分提升33%-62%，接近真人交互水平。
- 适合需要高鲁棒性测试的大模型代理训练与评估场景。
大语言模型（LLM）代理在与各类用户互动时面临真实用户行为多变、不合作等挑战，但大规模真实交互数据收集成本高昂。当前依赖的LLM用户模拟器行为同质且过于配合，导致代理在真实场景中表现不佳。为此，本文提出Persona Policies（PPol），一种即插即用的控制层，可在保持任务目标的前提下引入真实用户行为多样性。通过将角色生成建模为基于LLM的进化程序搜索，优化一个Python生成器以发现并转化为任务兼容的角色扮演策略。候选生成器由多目标适应度分数指导，兼顾人类相似度与行为模式覆盖率。优化后生成器可为任意任务域生成多样化的类人用户。在tau^2-bench的零售与航空领域测试中，PPol模拟器的适应度得分比基线提升33%-62%。盲评中，标注者认为PPol模拟用户为真人达80.4%，接近真实人类轨迹，几乎是基线模拟器的两倍。使用PPol训练的代理对分布外挑战更具鲁棒性，任务成功率相对提升17%。该方法无需改变任务或奖励即可增强基于模拟器的评估与训练。
原文摘要 · Abstract (English)
Large Language Model (LLM) agents are increasingly deployed in settings where they interact with a wide variety of people, including users who are unclear, impatient, or reluctant to share information. However, collecting real interaction data at scale remains expensive. The field has turned to LLM-based user simulators as stand-ins, but these simulators inherit the behavior of their underlying models: cooperative and homogeneous. As a result, agents that appear strong in simulation often fail under the unseen, diverse communication patterns of real users. To narrow this gap, we introduce Persona Policies (PPol), a plug-and-play control layer that induces realistic behavioral variation in user simulators while preserving the original task goals. Rather than hand-crafting personas, we cast persona generation as an LLM-driven evolutionary program search that optimizes a Python generator to discover behaviors and translate them into task-preserving roleplay policies. Candidate generators are guided by a multi-objective fitness score combining human-likeness with broad coverage of human behavioral patterns. Once optimized, the generator produces a diverse population of human-like personas for any task in the domain. Across tau^2-bench retail and airline domains, evolved PPol programs yield 33-62% absolute gains in fitness score over the baseline simulator. In a blinded evaluation, annotators rated PPol-conditioned users as human 80.4% of the time, close to real human traces and nearly twice as frequently as baseline simulators. Agents trained with PPol are more robust to challenging, out-of-distribution behaviors, improving task success by +17% relative to training only on existing simulated interactions. This offers a novel approach to strengthen simulator-based evaluation and training without changing tasks or rewards.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。