PERM: Psychology-grounded Empathetic Reward Modeling for Large Language Models
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2026
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866914259283214336 |
|---|---|
| author | Wang, Chengbing Zheng, Wuqiang Zhang, Yang Zhu, Fengbin Cheng, Junyi Xie, Yi Wang, Wenjie Feng, Fuli |
| author_facet | Wang, Chengbing Zheng, Wuqiang Zhang, Yang Zhu, Fengbin Cheng, Junyi Xie, Yi Wang, Wenjie Feng, Fuli |
| contents | Large Language Models (LLMs) are increasingly deployed in human-centric applications, yet they often fail to provide substantive emotional support. While Reinforcement Learning (RL) has been utilized to enhance empathy of LLMs, existing reward models typically evaluate empathy from a single perspective, overlooking the inherently bidirectional interaction nature of empathy between the supporter and seeker as defined by Empathy Cycle theory. To address this limitation, we propose Psychology-grounded Empathetic Reward Modeling (PERM). PERM operationalizes empathy evaluation through a bidirectional decomposition: 1) Supporter perspective, assessing internal resonation and communicative expression; 2) Seeker perspective, evaluating emotional reception. Additionally, it incorporates a bystander perspective to monitor overall interaction quality. Extensive experiments on a widely-used emotional intelligence benchmark and an industrial daily conversation dataset demonstrate that PERM outperforms state-of-the-art baselines by over 10\%. Furthermore, a blinded user study reveals a 70\% preference for our approach, highlighting its efficacy in generating more empathetic responses. Our code, dataset, and models are available at https://github.com/ZhengWwwq/PERM. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2601_10532 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | PERM: Psychology-grounded Empathetic Reward Modeling for Large Language Models Wang, Chengbing Zheng, Wuqiang Zhang, Yang Zhu, Fengbin Cheng, Junyi Xie, Yi Wang, Wenjie Feng, Fuli Computation and Language Large Language Models (LLMs) are increasingly deployed in human-centric applications, yet they often fail to provide substantive emotional support. While Reinforcement Learning (RL) has been utilized to enhance empathy of LLMs, existing reward models typically evaluate empathy from a single perspective, overlooking the inherently bidirectional interaction nature of empathy between the supporter and seeker as defined by Empathy Cycle theory. To address this limitation, we propose Psychology-grounded Empathetic Reward Modeling (PERM). PERM operationalizes empathy evaluation through a bidirectional decomposition: 1) Supporter perspective, assessing internal resonation and communicative expression; 2) Seeker perspective, evaluating emotional reception. Additionally, it incorporates a bystander perspective to monitor overall interaction quality. Extensive experiments on a widely-used emotional intelligence benchmark and an industrial daily conversation dataset demonstrate that PERM outperforms state-of-the-art baselines by over 10\%. Furthermore, a blinded user study reveals a 70\% preference for our approach, highlighting its efficacy in generating more empathetic responses. Our code, dataset, and models are available at https://github.com/ZhengWwwq/PERM. |
| title | PERM: Psychology-grounded Empathetic Reward Modeling for Large Language Models |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2601.10532 |