Distilling Instructions into Weights: Closed-Form Edits Fix a Robot Policy's Misreading
cs.RO, cs.AI, cs.LG
Submitted: 2026-11-09
Updated: 2026-11-09
Code: https://github.com/adeeb10abbas/weightsmith
Project page: https://ali-adeeb.com/weightsmith
Terminology
Sources
- $\pi_{0.5}$: a Vision-Language-Action Model with Open-World Generalization
- LIBERO-Para: A Diagnostic Benchmark and Metrics for Paraphrase Robustness in VLA Models
- Same Goal, Different Words: Spatial Instruction Sensitivity in World-Action Models
- PaliGemma: A versatile 3B VLM for transfer
- $\pi_0$: A Vision-Language-Action Flow Model for General Robot Control
- Learning by Distilling Context
- Steering Language Models With Activation Engineering
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