Research map / Large language models / Agents, code and math
MCTSr-Zero: Self-Reflective Psychological Counseling Dialogues Generation via Principles and Adaptive Exploration
Hao Lu, Yanchi Gu, Haoyuan Huang, Yulin Zhou, Ningxin Zhu, Chen Li
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add 1 iteration of Self-Refine (ablation) | gpt-4.1-mini · PsyEval | PsyEval score: 83.60 -> 86.39 | Improved |
| apply MCTSr-Zero for 4 iterations (ablation) | gpt-4.1-mini · PsyEval | PsyEval score: 83.60 -> 90.18 | Improved |
| use full MCTSr-Zero (meta-prompt adaptation and standards) instead of the w/o meta;standards variant (ablation) | gpt-4.1-mini · PsyEval | PsyEval score | Improved |
| increase refinement iterations (ablation) | refinement methods on gpt-4.1-mini · PsyEval | PsyEval score | Improved |
| fine-tune GLM-4-32B-0414 on MCTSr-Zero-Psy dialogues (SFT + SimPO) | GLM-4-32B-0414 · PsyEval | total score: 80.92 -> 90.93 | Improved |
| fine-tune GLM-4-9B-0414 on MCTSr-Zero-Psy dialogues (SFT + SimPO) | GLM-4-9B-0414 · PsyEval | total score: 75.74 -> 90.72 | Improved |
Closest papers by meaning
- STAMPsy: Towards SpatioTemporal-Aware Mixed-Type Dialogues for Psychological Counseling (AAAI 2025 · similarity 0.89)
- An LLM-based Simulation Framework for Embodied Conversational Agents in Psychological Counseling (AAAI 2026 · similarity 0.87)
- EmpathyChat: Structured Cognitive Reasoning in Empathetic Spoken Dialogue (NeurIPS 2026 · similarity 0.85)
- TherapyGym: Evaluating and Aligning Clinical Fidelity and Safety in Therapy Chatbots (NeurIPS 2026 · similarity 0.85)
- SocialSim: Towards Socialized Simulation of Emotional Support Conversation (AAAI 2025 · similarity 0.84)
- RLVER: Reinforcement Learning with Verifiable Emotion Rewards for Empathetic Agents (ICLR 2026 · similarity 0.84)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
