Suppression Sticks, Locality Is Fragile: A Closed-Loop Target-and-Control Audit of Task-Vector Negation in VLA Policies
Shaoguang Wang, Weiyu Guo, Rushi Dai, Yiren Zhao, Yandong Guo, Hui Xiong
cs.RO, cs.LG
Submitted: 2026-08-05
Comments: 28 pages, 14 figures, 40 tables. Preprint
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Terminology
Sources
- $\pi_0$: A Vision-Language-Action Flow Model for General Robot Control
- PHASER: Phase-Aware and Semantic Experience Replay for Vision-Language-Action Models
- Do Unlearning Methods Remove Information from Language Model Weights?
- Who's Harry Potter? Approximate Unlearning in LLMs
- MergeVLA: Cross-Skill Model Merging Toward a Generalist Vision-Language-Action Agent
- NeuroVLA: A Brain-inspired Embodied Intelligence for Fluid and Fast Reflexive Robotics Control
- RECALL: Recovery Experience Collection for Active Lifelong Learning in Vision-Language-Action Models
- OpenVLA: An Open-Source Vision-Language-Action Model
- Knowledge Vector Weakening: Efficient Training-free Unlearning for Large Vision-Language Models
- Spatial Memory for Out-of-Vision Manipulation in Vision-Language-Action
- CATA: Continual Machine Unlearning via Conflict-Averse Task Arithmetic
- Pretrained Vision-Language-Action Models are Surprisingly Resistant to Forgetting in Continual Learning
- An Adversarial Perspective on Machine Unlearning for AI Safety
- Eight Methods to Evaluate Robust Unlearning in LLMs
- TOFU: A Task of Fictitious Unlearning for LLMs
- $\pi_{0.5}$: a Vision-Language-Action Model with Open-World Generalization
- VLA-Forget: Vision-Language-Action Unlearning for Embodied Foundation Models
- MUSE: Machine Unlearning Six-Way Evaluation for Language Models
- SmolVLA: A Vision-Language-Action Model for Affordable and Efficient Robotics
- Where to Focus: Query-Modulated Multimodal Keyframe Selection for Long Video Understanding
Related papers
- FMT x: An Efficient and Asymptotically Optimal Extension of the Fast Marching Tree for Dynamic Replanning
- MPCFormer: A physics-informed data-driven approach for explainable socially-aware autonomous driving
- RoboLab: A High-Fidelity Simulation Benchmark for Analysis of Task Generalist Policies
- ARC: A Reasoning Recipe for Robot Foundation Models
- HRDexDB: A 4D Dexterous Grasping Dataset Across Human and Multiple Robot Embodiments
- APT: Action Expert Pretraining Improves Instruction Generalization of Vision-Language-Action Policies