{"ID":2866375,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2510.02320","arxiv_id":"2510.02320","title":"WEE-Therapy: A Mixture of Weak Encoders Framework for Psychological Counseling Dialogue Analysis","abstract":"The advancement of computational psychology requires AI tools capable of deeply understanding counseling dialogues. Existing audio language models (AudioLLMs) often rely on single speech encoders pre-trained on general data, struggling to capture domain-specific features like complex emotions and professional techniques. To address this, we propose WEE-Therapy, a multi-task AudioLLM incorporating a Weak Encoder Ensemble (WEE) mechanism. This supplements a powerful base encoder with a pool of lightweight, specialized encoders. A novel dual-routing strategy combines stable, data-independent domain knowledge with dynamic, data-dependent expert selection. Evaluated on emotion recognition, technique classification, risk detection, and summarization, WEE-Therapy achieves significant performance gains across all tasks with minimal parameter overhead, demonstrating strong potential for AI-assisted clinical analysis.","short_abstract":"The advancement of computational psychology requires AI tools capable of deeply understanding counseling dialogues. Existing audio language models (AudioLLMs) often rely on single speech encoders pre-trained on general data, struggling to capture domain-specific features like complex emotions and professional technique...","url_abs":"https://arxiv.org/abs/2510.02320","url_pdf":"https://arxiv.org/pdf/2510.02320v1","authors":"[\"Yongqi Kang\",\"Yong Zhao\"]","published":"2025-09-24T05:43:53Z","proceeding":"eess.AS","tasks":"[\"eess.AS\",\"cs.CL\",\"cs.LG\",\"cs.SD\"]","methods":"[\"Large Language Model\",\"Language Model\"]","has_code":false}