{"ID":2886961,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2508.02458","arxiv_id":"2508.02458","title":"From Stimuli to Minds: Enhancing Psychological Reasoning in LLMs via Bilateral Reinforcement Learning","abstract":"Large Language Models show promise in emotion understanding, social reasoning, and empathy, yet they struggle with psychologically grounded tasks that require inferring implicit mental states in context-rich, ambiguous settings. These limitations arise from the absence of theory-aligned supervision and the difficulty of capturing nuanced mental processes in real-world narratives. To address this gap, we leverage expert-labeled, psychologically rich scenarios and propose a trajectory-aware reinforcement learning framework that explicitly imitates expert psychological thought patterns. By integrating real-world stimuli with structured reasoning guidance, our approach enables compact models to internalize social-cognitive principles, perform nuanced psychological inference, and support continual self-improvement. Comprehensive experiments across multiple benchmarks further demonstrate that our models achieve expert-level interpretive capabilities, exhibiting strong out-of-distribution generalization and robust continual learning across diverse, challenging, and psychologically grounded tasks.","short_abstract":"Large Language Models show promise in emotion understanding, social reasoning, and empathy, yet they struggle with psychologically grounded tasks that require inferring implicit mental states in context-rich, ambiguous settings. These limitations arise from the absence of theory-aligned supervision and the difficulty o...","url_abs":"https://arxiv.org/abs/2508.02458","url_pdf":"https://arxiv.org/pdf/2508.02458v4","authors":"[\"Yichao Feng\",\"Haoran Luo\",\"Lang Feng\",\"Shuai Zhao\",\"Anh Tuan Luu\"]","published":"2025-08-04T14:24:30Z","proceeding":"cs.DB","tasks":"[\"cs.DB\"]","methods":"[\"Reinforcement Learning\",\"Large Language Model\",\"Language Model\"]","has_code":false}