Robo-Dopamine: General Process Reward Modeling for High-Precision Robotic Manipulation
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866917174275211264 |
|---|---|
| author | Tan, Huajie Chen, Sixiang Xu, Yijie Wang, Zixiao Ji, Yuheng Chi, Cheng Lyu, Yaoxu Zhao, Zhongxia Chen, Xiansheng Co, Peterson Xie, Shaoxuan Yao, Guocai Wang, Pengwei Wang, Zhongyuan Zhang, Shanghang |
| author_facet | Tan, Huajie Chen, Sixiang Xu, Yijie Wang, Zixiao Ji, Yuheng Chi, Cheng Lyu, Yaoxu Zhao, Zhongxia Chen, Xiansheng Co, Peterson Xie, Shaoxuan Yao, Guocai Wang, Pengwei Wang, Zhongyuan Zhang, Shanghang |
| contents | The primary obstacle for applying reinforcement learning (RL) to real-world robotics is the design of effective reward functions. While recently learning-based Process Reward Models (PRMs) are a promising direction, they are often hindered by two fundamental limitations: their reward models lack step-aware understanding and rely on single-view perception, leading to unreliable assessments of fine-grained manipulation progress; and their reward shaping procedures are theoretically unsound, often inducing a semantic trap that misguides policy optimization. To address these, we introduce Dopamine-Reward, a novel reward modeling method for learning a general-purpose, step-aware process reward model from multi-view inputs. At its core is our General Reward Model (GRM), trained on a vast 3,400+ hour dataset, which leverages Step-wise Reward Discretization for structural understanding and Multi-Perspective Reward Fusion to overcome perceptual limitations. Building upon Dopamine-Reward, we propose Dopamine-RL, a robust policy learning framework that employs a theoretically-sound Policy-Invariant Reward Shaping method, which enables the agent to leverage dense rewards for efficient self-improvement without altering the optimal policy, thereby fundamentally avoiding the semantic trap. Extensive experiments across diverse simulated and real-world tasks validate our approach. GRM achieves state-of-the-art accuracy in reward assessment, and Dopamine-RL built on GRM significantly improves policy learning efficiency. For instance, after GRM is adapted to a new task in a one-shot manner from a single expert trajectory, the resulting reward model enables Dopamine-RL to improve the policy from near-zero to 95% success with only 150 online rollouts (approximately 1 hour of real robot interaction), while retaining strong generalization across tasks. Project website: https://robo-dopamine.github.io |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2512_23703 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Robo-Dopamine: General Process Reward Modeling for High-Precision Robotic Manipulation Tan, Huajie Chen, Sixiang Xu, Yijie Wang, Zixiao Ji, Yuheng Chi, Cheng Lyu, Yaoxu Zhao, Zhongxia Chen, Xiansheng Co, Peterson Xie, Shaoxuan Yao, Guocai Wang, Pengwei Wang, Zhongyuan Zhang, Shanghang Robotics The primary obstacle for applying reinforcement learning (RL) to real-world robotics is the design of effective reward functions. While recently learning-based Process Reward Models (PRMs) are a promising direction, they are often hindered by two fundamental limitations: their reward models lack step-aware understanding and rely on single-view perception, leading to unreliable assessments of fine-grained manipulation progress; and their reward shaping procedures are theoretically unsound, often inducing a semantic trap that misguides policy optimization. To address these, we introduce Dopamine-Reward, a novel reward modeling method for learning a general-purpose, step-aware process reward model from multi-view inputs. At its core is our General Reward Model (GRM), trained on a vast 3,400+ hour dataset, which leverages Step-wise Reward Discretization for structural understanding and Multi-Perspective Reward Fusion to overcome perceptual limitations. Building upon Dopamine-Reward, we propose Dopamine-RL, a robust policy learning framework that employs a theoretically-sound Policy-Invariant Reward Shaping method, which enables the agent to leverage dense rewards for efficient self-improvement without altering the optimal policy, thereby fundamentally avoiding the semantic trap. Extensive experiments across diverse simulated and real-world tasks validate our approach. GRM achieves state-of-the-art accuracy in reward assessment, and Dopamine-RL built on GRM significantly improves policy learning efficiency. For instance, after GRM is adapted to a new task in a one-shot manner from a single expert trajectory, the resulting reward model enables Dopamine-RL to improve the policy from near-zero to 95% success with only 150 online rollouts (approximately 1 hour of real robot interaction), while retaining strong generalization across tasks. Project website: https://robo-dopamine.github.io |
| title | Robo-Dopamine: General Process Reward Modeling for High-Precision Robotic Manipulation |
| topic | Robotics |
| url | https://arxiv.org/abs/2512.23703 |