CARE: An Explainable Computational Framework for Assessing Client-Perceived Therapeutic Alliance Using Large Language Models

2026-02-24

AI Summary

This study addresses the challenge of accurately and promptly capturing clients’ multidimensional perceptions of the therapeutic alliance—a task where existing methods often fall short in both performance and interpretability due to inadequate contextual modeling. To bridge this gap, we propose an interpretable prediction framework grounded in the LLaMA-3.1-8B-Instruct large language model, introducing a novel rationale-augmented supervision mechanism. Our approach leverages the CounselingWAI dataset, fine-tuned with 9,516 expert-annotated rationales, and achieves a Pearson correlation coefficient improvement exceeding 70% in predicting client ratings. This substantial gain significantly narrows the discrepancy between therapist assessments and clients’ actual perceptions. Furthermore, the model delivers highly interpretable and actionable clinical insights, demonstrating strong practical utility in real-world Chinese psychotherapy settings.

Technology Category

• Cognitive Modeling & Cognitive Systems: Conceptual Inference and Reasoning
• Natural Language Processing: Interpretability, Analysis, and Evaluation of NLP Models
• Machine Learning: Large Multimodal Models (LMMs)

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions

Abstract

Client perceptions of the therapeutic alliance are critical for counseling effectiveness. Accurately capturing these perceptions remains challenging, as traditional post-session questionnaires are burdensome and often delayed, while existing computational approaches produce coarse scores, lack interpretable rationales, and fail to model holistic session context. We present CARE, an LLM-based framework to automatically predict multi-dimensional alliance scores and generate interpretable rationales from counseling transcripts. Built on the CounselingWAI dataset and enriched with 9,516 expert-curated rationales, CARE is fine-tuned using rationale-augmented supervision with the LLaMA-3.1-8B-Instruct backbone. Experiments show that CARE outperforms leading LLMs and substantially reduces the gap between counselor evaluations and client-perceived alliance, achieving over 70% higher Pearson correlation with client ratings. Rationale-augmented supervision further improves predictive accuracy. CARE also produces high-quality, contextually grounded rationales, validated by both automatic and human evaluations. Applied to real-world Chinese online counseling sessions, CARE uncovers common alliance-building challenges, illustrates how interaction patterns shape alliance development, and provides actionable insights, demonstrating its potential as an AI-assisted tool for supporting mental health care.

Problem

• therapeutic alliance
• client perception
• explainable AI
• large language models
• counseling

Innovation

• explainable AI
• therapeutic alliance
• large language models
• rationale-augmented supervision
• mental health counseling