Robo-Dopamine: General Process Reward Modeling for High-Precision Robotic Manipulation

TL;DR

Robo-Dopamine tackles the core challenge of reward design for real-world robotic RL by introducing a General Reward Model (GRM) trained on a large, multi-view dataset to provide step-aware, fine-grained progress signals. Built on GRM, Dopamine-RL employs a theoretically grounded Policy-Invariant Reward Shaping scheme that converts dense progress signals into a shaping term while preserving the optimal policy, enabling rapid policy learning with strong generalization. The approach achieves state-of-the-art progress accuracy (GRM > 92.8% and VOC ≈ 0.953), one-shot adaptation from a single demonstration to 95% success within roughly 150 online rollouts, and robust performance across unseen layouts and object variations. This yields a scalable, generalizable framework for dense reward modeling and efficient real-world manipulation, with potential extensions to continuous video reasoning and multi-modal sensing.

Abstract

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

Figures (11)

• Figure 1: Overview of Robo-Dopamine. Robo-Dopamine integrates large-scale reward modeling with a robust policy learning algorithm. (Left) We construct a General Reward Model (GRM) trained on a large and diverse 35M-sample dataset spanning real-world, simulation, and human-centric videos with our Dopamine-Reward, a step-aware fine-grained reward modeling method. This GRM learns to predict fine-grained, relative progress between states to accurately assess task progression. (Bottom Right) The pre-trained GRM is adapted to new tasks and provides dense reward signals to our Dopamine-RL framework. By using a theoretically-sound Policy-Invariant Reward Shaping method, Dopamine-RL efficiently guides the policy during online interactions without misaligning the task objective. (Top Right) Our integrated approach establishes a new state-of-the-art in reward accuracy (radar chart) and demonstrates high training efficiency, significantly boosting policy success rates in both simulation and the real world (bar chart).
• Figure 2: The overview of our method. Our framework is composed of two core components: (a) Dopamine-Reward Modeling Method and (b) Dopamine-RL Training Framework. (a) At the heart of our reward modeling is to build the General Reward Model (GRM), a vision-language model that is prompted with a task description and conditioned on multi-view images of initial, goal, "before," and "after" states to predict a relative progress or regress hop. To ensure a stable and accurate signal, we employ Multi-Perspective Progress Fusion, which combines incremental, forward-anchored, and backward-anchored predictions into a final fused reward. (b) The Dopamine-RL framework first adapts the pre-trained GRM to a novel task using a single demonstration (One-Shot GRM Adaptation). Subsequently, it uses a theoretically-sound Policy-Invariant Reward Shaping method to convert the GRM's dense output into a reward signal that accelerates learning without altering the optimal policy. This approach is universally compatible with a wide range of RL algorithms.
• Figure 3: Reward profiles on a challenging real-world rollout. We plot the reference reward from human annotations, the VLAC baseline, and our GRM along the same trajectory. Our GRM tracks the reference signal more faithfully, sharply penalizing incorrect insertions, low positions, and misalignments, and only assigning high reward near successful task completion.
• Figure 4: Real-world tasks and hardware setup. Left: eight representative long-horizon manipulation tasks used to evaluate Dopamine-Reward and Dopamine-RL, including insertion, circuit completion, folding, pick-and-place, and assembly tasks. Right: our multi-view hardware platform with the Pika teleoperation system and calibrated ZED cameras, providing synchronized wrist and third-person observations for GRM training and policy learning.
• Figure 5: Overview of GRM training data.(Left) The hierarchical composition of our 35M-sample training corpus. The dataset is derived from episodes spanning Real-World Robotics, Simulation, and Human-Centric domains, and is further expanded via multi-view augmentation. (Right) The long-tail distribution of task categories sorted by episode count (log scale). The dataset covers a broad spectrum of manipulation skills, ranging from atomic primitives (e.g., pick, push) to complex, multi-stage horizons (e.g., assemble, fold).