[
    {
        "id": "osp-19405",
        "type": "article-journal",
        "title": "PB-GRPO: Learning Socially Adaptive LLM Agents from Persona-Driven Simulation with Preference-Batched GRPO",
        "author": [
            {
                "family": "Wang",
                "given": "Jingquan"
            },
            {
                "family": "Yin",
                "given": "Jun"
            },
            {
                "family": "Han",
                "given": "Xu"
            },
            {
                "family": "Mei",
                "given": "Yongsheng"
            },
            {
                "family": "Hao",
                "given": "Jie"
            },
            {
                "family": "Guo",
                "given": "Bin"
            }
        ],
        "URL": "https://omanscience.com/en/articles/pb-grpo-learning-socially-adaptive-llm-agents-from-persona-driven-simulation-with-preference-batched-grpo",
        "language": "en",
        "issued": {
            "date-parts": [
                [
                    2026
                ]
            ]
        },
        "abstract": "Building LLMs that behave well socially, not merely correctly, requires Building LLMs that behave well socially, not merely correctly, requires more than producing locally helpful responses. A socially competent agent must infer users' unstated goals, respect their preferences, and adapt as the conversation unfolds. These behaviors are inherently multi-turn and social, making them hard to optimize: real interaction data is scarce, and user preferences are typically latent rather than directly observable. To address these challenges, we build on a persona-driven social simulation environment (consisting of a persona library, LLM-based user simulators, and a user-satisfaction scoring system ranging from [0, 1]), to introduce preference-batched GRPO (PB-GRPO), a post-training algorithm that learns socially adaptive policies from conversation-level feedback. Compared to vanilla GRPO, PB-GRPO computes advantages using a normalization estimated across a bucket of users with similar preferences, stabilizing training across a diverse social population. Empirical evidence shows that PB-GRPO improves models' social behavior over strong reinforcement learning baselines in our simulated environment."
    }
]