New CounselPlex Model Enhances Full-Duplex Counseling with Stressor Awareness

Researchers have developed CounselPlex, a novel full-duplex speech model designed for counseling applications. This framework addresses "Stressor Blindness," a common issue where models lose track of a user's core problem by focusing only on recent utterances. CounselPlex incorporates a stressor state that is updated during silences and injected into the model's input, alongside a training technique called RoPE Position Shift Augmentation (RPSA) to improve performance with longer conversational turns. Evaluations using an LLM judge indicated that CounselPlex offers superior counseling quality and better grounding of the user's stressor compared to existing strong baseline models. AI

IMPACT This research could lead to more empathetic and effective AI-driven counseling tools by improving conversational memory and context retention.

RANK_REASON The cluster contains a research paper detailing a new model and evaluation methodology. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.CL →

• alphaXiv
• arXiv
• CatalyzeX
• Connected Papers
• CounselPlex
• DagsHub
• Gotit.pub
• Hugging Face
• Litmaps
• LLM Judge
• RoPE Position Shift Augmentation
• ScienceCast
• scite Smart Citations

• paper
• model release

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →