GSBR: Geometric Support Behavior Regularization for Online RL
cs.RO, cs.LG
Submitted: 2026-11-03
Updated: 2026-11-03
Project page: https://charlesxu0124.github.io/gsbr
Terminology
Sources
- $\pi_0$: A Vision-Language-Action Flow Model for General Robot Control
- Randomized Ensembled Double Q-Learning: Learning Fast Without a Model
- ConRFT: A Reinforced Fine-tuning Method for VLA Models via Consistency Policy
- EXPO-FT: Sample-Efficient Reinforcement Learning Finetuning for Vision-Language-Action Models
- EXPO: Stable Reinforcement Learning with Expressive Policies
- Diffusion Actor-Critic: Formulating Constrained Policy Iteration as Diffusion Noise Regression for Offline Reinforcement Learning
- A Minimalist Approach to Offline Reinforcement Learning
- IDQL: Implicit Q-Learning as an Actor-Critic Method with Diffusion Policies
- Dropout Q-Functions for Doubly Efficient Reinforcement Learning
- Imitation Bootstrapped Reinforcement Learning
- Residual Reinforcement Learning for Robot Control
- Stabilizing Off-Policy Q-Learning via Bootstrapping Error Reduction
- Flow Matching for Generative Modeling
- VLA-RL: Towards Masterful and General Robotic Manipulation with Scalable Reinforcement Learning
- SERL: A Software Suite for Sample-Efficient Robotic Reinforcement Learning
- Precise and Dexterous Robotic Manipulation via Human-in-the-Loop Reinforcement Learning
- Policy Agnostic RL: Offline RL and Online RL Fine-Tuning of Any Class and Backbone
- Behavioral Cloning Mystery
- $\pi^{*}_{0.6}$: a VLA That Learns From Experience
- Diffusion Policy Policy Optimization
Related papers
- Rethinking Actions: Continuity-Aware Priors for World-Model Planning
- MOSAIC: Modular Obstacle and Spatial-Aware Framework for Manipulation In Cluttered Scenes
- SAGAS: Semantic-Aware Graph-Assisted Stitching for Offline Temporal Logic Planning
- DexSeed: Expanding Human Hand-Object Interactions into Robot Demonstrations for Dexterous Policy Learning
- LaST0.5: A Generalist World Action Model with Interaction-Centric Latent Reasoning
- CIALLO: Learning Shared Intent Representations for Cross-Embodiment Manipulation