{"ID":3053352,"CreatedAt":"2026-06-04T04:41:36.695875263Z","UpdatedAt":"2026-06-06T02:58:33.341803073Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2606.04389","arxiv_id":"2606.04389","title":"When Clients Stop Following: A Cognitive Conceptualization Diagram-driven Framework for Strategic Counseling","abstract":"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.","short_abstract":"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 thera...","url_abs":"https://arxiv.org/abs/2606.04389","url_pdf":"https://arxiv.org/pdf/2606.04389v1","authors":"[\"Yihao Qin\",\"Junyi Zhao\",\"Changsheng Ma\",\"Yongfeng Tao\",\"Minqiang Yang\",\"Chang Liu\",\"Bin Hu\"]","published":"2026-06-03T03:09:10Z","proceeding":"cs.CL","tasks":"[\"cs.CL\"]","methods":"[\"Reinforcement Learning\",\"Large Language Model\",\"Language Model\"]","has_code":false}