Off-Policy RL with Online Inference Tuning: My Solution to the CoRL 2026 RoboEval Challenge
cs.RO, cs.CV, cs.LG
Submitted: 2026-11-08
Updated: 2026-11-08
Terminology
Sources
- RoboEval: Where Robotic Manipulation Meets Structured and Scalable Evaluation
- Learning to Fold: prizewinning solution at LeHome Challenge 2026 (1st place online, 2nd offline)
- $\pi_{0.5}$: a Vision-Language-Action Model with Open-World Generalization
- Task adaptation of Vision-Language-Action model: 1st Place Solution for the 2025 BEHAVIOR Challenge
- High-Dimensional Continuous Control Using Generalized Advantage Estimation
- Advantage-Weighted Regression: Simple and Scalable Off-Policy Reinforcement Learning
- $\pi^{*}_{0.6}$: a VLA That Learns From Experience
- Classifier-Free Diffusion Guidance
- A Tutorial on Thompson Sampling
- PaliGemma: A versatile 3B VLM for transfer
- FAST: Efficient Action Tokenization for Vision-Language-Action Models
- Real-Time Execution of Action Chunking Flow Policies
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