analiza aiqbench
This research introduces ESCRAG-R1, a framework that merges retrieval-augmented generation with Group Relative Policy Optimization to enhance emotional support systems. By integrating external psychological knowledge directly into the reinforcement learning loop, the model achieves a more natural alignment between clinical expertise and empathetic expression without relying on superficial data splicing. The accompanying ESC-Preference dataset provides high-quality, empathy-aware reward signals essential for training robust therapeutic AI agents.
ESCRAG-R1: Retrieval-Augmented Reinforcement Learning for Emotional Support Conversation
arXiv:2608.21925v1 Announce Type: new Abstract: Emotional Support Conversation (ESC) systems aim to provide holistic support by balancing professional therapeutic competence with natural empathy. However, existing methods struggle to simultaneously achieve structured, stage-aware reasoning and seamless empathy-expertise alignment, often resulting in an artificial splicing of clinical strategies and generic reassurance. To overcome these limitations, we propose ESCRAG-R1, a unified framework that integrates retrieval-based psychological guidance into Group Relative Policy Optimization (GRPO). By incorporating retrieval into the reinforcement learning loop, ESCRAG-R1 transforms external knowledge into a robust learning signal that stimulates explicit internal reasoning prior to generation and fundamentally reshapes the model's internal policy. To provide the reliable supervision required for this optimization, we construct ESC-Preference, a high-quality dataset based on a Client--Counselor--Judge evaluation framework that delivers precise, empathy-aware reward signals. Extensive experiments demonstrate that ESCRAG-R1 significantly outperforms existing baselines by mitigating superficial splicing and realizing a natural integration of professional guidance and empathetic expression. Code and datasets are released at https://github.com/Matcha-Liu/ESCRAG-R1.
Abstract:Emotional Support Conversation (ESC) systems aim to provide holistic support by balancing professional therapeutic competence with natural empathy. However, existing methods struggle to simultaneously achieve structured, stage-aware reasoning and seamless empathy-expertise alignment, often resulting in an artificial splicing of clinical strategies and generic reassurance. To overcome these limitations, we propose ESCRAG-R1, a unified framework that integrates retrieval-based psychological guidance into Group Relative Policy Optimization (GRPO). By incorporating retrieval into the reinforcement learning loop, ESCRAG-R1 transforms external knowledge into a robust learning signal that stimulates explicit internal reasoning prior to generation and fundamentally reshapes the model's internal policy. To provide the reliable supervision required for this optimization, we construct ESC-Preference, a high-quality dataset based on a Client--Counselor--Judge evaluation framework that delivers precise, empathy-aware reward signals. Extensive experiments demonstrate that ESCRAG-R1 significantly outperforms existing baselines by mitigating superficial splicing and realizing a natural integration of professional guidance and empathetic expression. Code and datasets are released at this https URL.
Artificial Intelligence (cs.AI)
arXiv:2608.21925 [cs.AI]
(or arXiv:2608.21925v1 [cs.AI] for this version)
https://doi.org/10.48550/arXiv.2608.21925 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Text integral sub licența deschisă a publicației · 251 cuvinte
Subiecte conexe
AI customer service: strategy, agents, and solutions guide
Databricks · 27 iul. 2026
Introducing Instella-MoE: A State-of-the-Art Fully Open Mixture-of-Experts Language Model
AMD · 24 iul. 2026
How I Turned AI to the Dark Side
IEEE Spectrum · 14 iul. 2026
What are AI Hallucinations?
Databricks · 13 aug. 2026