PAIR: Proprioception-Aligned Image-Ray Representations for Viewpoint-Robust Manipulation
cs.RO
Submitted: 2026-11-09
Updated: 2026-11-09
Terminology
Sources
- Multi-Camera View Scaling for Data-Efficient Robot Imitation Learning
- SCoPE: Sightline-Coordinate Positional Encoding for Video Diffusion Transformers
- Seeing to Act, Prompting to Specify: A Bayesian Factorization of Vision Language Action Policy
- RayViT: Ray-Conditioned Visual Representations for Viewpoint-Robust Imitation Learning
- SigLIP 2: Multilingual Vision-Language Encoders with Improved Semantic Understanding, Localization, and Dense Features
- MimicGen: A Data Generation System for Scalable Robot Learning using Human Demonstrations
- Learning Fine-Grained Bimanual Manipulation with Low-Cost Hardware
- SmolVLA: A Vision-Language-Action Model for Affordable and Efficient Robotics
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