From Stimuli to Minds: Enhancing Psychological Reasoning in LLMs via Bilateral Reinforcement Learning

TL;DR

This paper introduces a reinforcement learning framework that improves large language models' ability to perform nuanced psychological reasoning by imitating expert mental state inference in complex, real-world scenarios.

Contribution

It presents a trajectory-aware reinforcement learning approach that integrates expert psychological reasoning with real-world stimuli to enhance LLMs' social-cognitive understanding.

Findings

Models achieve expert-level interpretive capabilities.

Strong out-of-distribution generalization.

Robust continual learning across tasks.

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.