Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2512.01282 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866908686277935104 |
|---|---|
| author | Yuan, Jiahao Cui, Zhiqing Wang, Hanqing Gao, Yuansheng Zhou, Yucheng Naseem, Usman |
| author_facet | Yuan, Jiahao Cui, Zhiqing Wang, Hanqing Gao, Yuansheng Zhou, Yucheng Naseem, Usman |
| contents | As web platforms evolve towards greater personalization and emotional complexity, conversational agents must transcend superficial empathy to demonstrate identity-aware emotional reasoning. However, existing systems face two limitations: (1) reliance on situation-centric datasets lacking persistent user identity, which hampers the capture of personalized affective nuances; and (2) dependence on opaque, coarse reward signals that hinder development of verifiable empathetic reasoning. To address these gaps, we introduce KardiaBench, a large-scale user-grounded benchmark comprising 178,080 QA pairs across 22,080 multi-turn conversations anchored to 671 real-world profiles. The dataset is constructed via a model-in-the-loop pipeline with iterative rubric-guided refinement to ensure psychological plausibility and persona consistency. This progressive empathy pipeline that integrates user comprehension, contextual reasoning, and emotion perception into conversations, followed by iterative critique and rubric-based refinement to ensure psychological plausibility, emotional fidelity, and persona consistency. Building on this, we propose Kardia-R1, a framework that trains models for interpretable, stepwise empathetic cognition. Kardia-R1 leverages Rubric-as-Judge Empathetic Reinforcement Learning (Rubric-ERL), a GRPO-based method that uses explainable, human-aligned rubric rewards to tightly couple user understanding, emotional inference, and supportive response generation. Extensive experiments across four LLM backbones demonstrate that Kardia-R1 consistently outperforms othet methods in emotion accuracy, empathy, relevance, persona consistency, and safety. Our dataset and model will be released at https://github.com/JhCircle/Kardia-R1. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2512_01282 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Kardia-R1: Unleashing LLMs to Reason toward Understanding and Empathy for Emotional Support via Rubric-as-Judge Reinforcement Learning Yuan, Jiahao Cui, Zhiqing Wang, Hanqing Gao, Yuansheng Zhou, Yucheng Naseem, Usman Computation and Language Artificial Intelligence As web platforms evolve towards greater personalization and emotional complexity, conversational agents must transcend superficial empathy to demonstrate identity-aware emotional reasoning. However, existing systems face two limitations: (1) reliance on situation-centric datasets lacking persistent user identity, which hampers the capture of personalized affective nuances; and (2) dependence on opaque, coarse reward signals that hinder development of verifiable empathetic reasoning. To address these gaps, we introduce KardiaBench, a large-scale user-grounded benchmark comprising 178,080 QA pairs across 22,080 multi-turn conversations anchored to 671 real-world profiles. The dataset is constructed via a model-in-the-loop pipeline with iterative rubric-guided refinement to ensure psychological plausibility and persona consistency. This progressive empathy pipeline that integrates user comprehension, contextual reasoning, and emotion perception into conversations, followed by iterative critique and rubric-based refinement to ensure psychological plausibility, emotional fidelity, and persona consistency. Building on this, we propose Kardia-R1, a framework that trains models for interpretable, stepwise empathetic cognition. Kardia-R1 leverages Rubric-as-Judge Empathetic Reinforcement Learning (Rubric-ERL), a GRPO-based method that uses explainable, human-aligned rubric rewards to tightly couple user understanding, emotional inference, and supportive response generation. Extensive experiments across four LLM backbones demonstrate that Kardia-R1 consistently outperforms othet methods in emotion accuracy, empathy, relevance, persona consistency, and safety. Our dataset and model will be released at https://github.com/JhCircle/Kardia-R1. |
| title | Kardia-R1: Unleashing LLMs to Reason toward Understanding and Empathy for Emotional Support via Rubric-as-Judge Reinforcement Learning |
| topic | Computation and Language Artificial Intelligence |
| url | https://arxiv.org/abs/2512.01282 |