Beyond Static Personas: Situational Personality Steering for Large Language Models

arXiv:2604.13846v1 Announce Type: new Abstract: Personalized Large Language Models (LLMs) facilitate more natural, human-like interactions in human-centric applications. However, existing personalization methods are constrained by limited controllability and high resource demands. Furthermore, their reliance on static personality modeling restricts adaptability across varying situations. To address these limitations, we first demonstrate the existence of situation-dependency and consistent situation-behavior patterns within LLM personalities through a multi-perspective analysis of persona neurons. Building on these insights, we propose IRIS, a training-free, neuron-based Identify-Retrieve-Steer framework for advanced situational personality steering. Our approach comprises situational persona neuron identification, situation-aware neuron retrieval, and similarity-weighted steering. We empirically validate our framework on PersonalityBench and our newly introduced SPBench, a comprehensive situational personality benchmark. Experimental results show that our method surpasses best-performing baselines, demonstrating IRIS's generalization and robustness to complex, unseen situations and different models architecture.

Related

• Towards Real-world Human Behavior Simulation: Benchmarking Large Language Models on Long-horizon, Cross-scenario, Heterogeneous Behavior Traces
• WASD: Locating Critical Neurons as Sufficient Conditions for Explaining and Controlling LLM Behavior
• PersonaVLM: Long-Term Personalized Multimodal LLMs
• Modeling Multi-Dimensional Cognitive States in Large Language Models under Cognitive Crowding

Source: arXiv cs.CL | 2026-04-16