When Clients Stop Following: A Cognitive Conceptualization Diagram-driven Framework for Strategic Counseling
For the field of AI-assisted psychological counseling, this work addresses the critical problem of evaluation mismatch caused by unrealistic client compliance, offering a more realistic benchmark and training method.
The paper identifies that existing LLM counseling benchmarks overestimate therapeutic progress due to highly cooperative simulated clients, and proposes a CBT-grounded framework with a resistance-aware client simulator and a dual-module strategic counseling system, demonstrating improved robustness under challenging interactions.
Large Language Models (LLMs) show promise in psychological counseling, yet existing benchmarks rely heavily on highly cooperative simulated clients. We observe a critical counselor-following phenomenon: these clients often rapidly shift from resistance to compliance after only a few turns, creating an illusion of therapeutic progress and inflating scores under current evaluation protocols through superficial empathy. To address this evaluation mismatch, we propose a Cognitive Behavioral Therapy (CBT)-grounded resistance-aware framework. We introduce CARS, a client simulator that explicitly models dynamic resistance via Cognitive Conceptualization Diagrams (CCDs). We present STREAMS, a dual-module framework that decouples strategic reasoning (Thinker) from response generation (Presenter) and optimizes it via reinforcement learning. We further propose EWTS-MI, an entropy-weighted metric for evaluating responsiveness under high-friction interactions. Experiments across resistant and non-resistant counseling settings validate our findings on evaluation mismatch and demonstrate the effectiveness of resistance-aware training for improving strategic robustness under challenging counseling interactions.