Research
Beyond Prompt: Fine-grained Simulation of Cognitively Impaired Standardized Patients via Stochastic Steering
arXiv:2604.12210v1 Announce Type: new Abstract: Simulating Standardized Patients with cognitive impairment offers a scalable and ethical solution for clinical training. However, existing methods rely
arXiv:2604.12210v1 Announce Type: new Abstract: Simulating Standardized Patients with cognitive impairment offers a scalable and ethical solution for clinical training. However, existing methods rely on discrete prompt engineering and fail to capture the heterogeneity of deficits across varying domains and severity levels. To address this limitation, we propose StsPatient for the fine-grained simulation of cognitively impaired patients. We innovatively capture domain-specific features by extracting steering vectors from contrastive pairs of instructions and responses. Furthermore, we introduce a Stochastic Token Modulation (STM) mechanism to regulate the intervention probability. STM enables precise control over impairment severity while mitigating the instability of conventional vector methods. Comprehensive experiments demonstrate that StsPatient significantly outperforms baselines in both clinical authenticity and severity controllability.
Related
- LETGAMES: An LLM-Powered Gamified Approach to Cognitive Training for Patients with Cognitive Impairment
- A Systematic Analysis of the Impact of Persona Steering on LLM Capabilities
- Cognitive Training for Language Models: Towards General Capabilities via Cross-Entropy Games
- From Attribution to Action: A Human-Centered Application of Activation Steering
- Scalable High-Recall Constraint-Satisfaction-Based Information Retrieval for Clinical Trials Matching
Source: arXiv cs.AI | 2026-04-15