Correct Robots Before They Make Mistakes: Proactive Human-in-the-Loop Intervention via Preference Learning
cs.RO
Submitted: 2026-11-03
Updated: 2026-11-03
Project page: https://intuitive-robots.github.io/PHIL-page
Terminology
Sources
- RaC: Robot Learning for Long-Horizon Tasks by Scaling Recovery and Correction
- Predictive Preference Learning from Human Interventions
- Uncertainty Comes for Free: Human-in-the-Loop Policies with Diffusion Models
- FlowDAgger: Human-in-the-Loop Adaptation of Generative Robot Policies in Latent Space
- FlowCorrect: Efficient Interactive Correction of Generative Flow Policies for Robotic Manipulation
- Goal-Conditioned Imitation Learning using Score-based Diffusion Policies
- Set-Supervised Diffusion Policy: Learning Action-Chunking Diffusion through Corrections
- ARCap: Collecting High-quality Human Demonstrations for Robot Learning with Augmented Reality Feedback
- IRIS: An Immersive Robot Interaction System
- VR-DAgger: Immersive VR for Dexterous Data Collection and Uncertainty-Guided On-Policy Correction
- X-VLA: Soft-Prompted Transformer as Scalable Cross-Embodiment Vision-Language-Action Model
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