Cover Image

Contrastive Conceptor Activation Steering (COAST): Unlocking Vision-Language-Action Models through Hidden States

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Miao, Miranda Muqing, Kim, Subin, Yang, Brandon, Ungar, Lyle
Format:	Preprint
Published:	2026
Subjects:	Robotics Artificial Intelligence Machine Learning
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

Closing the Confidence-Faithfulness Gap in Large Language Models
by: Miao, Miranda Muqing, et al.
Published: (2026)

Correctness-Optimized Residual Activation Lens (CORAL): Transferrable and Calibration-Aware Inference-Time Steering
by: Miao, Miranda Muqing, et al.
Published: (2026)

Conceptors for Semantic Steering
by: Triantafyllopoulos, Ilias, et al.
Published: (2026)

10 Open Challenges Steering the Future of Vision-Language-Action Models
by: Poria, Soujanya, et al.
Published: (2025)

Steering Vision-Language-Action Models as Anti-Exploration: A Test-Time Scaling Approach
by: Yang, Siyuan, et al.
Published: (2025)

MoRE: Unlocking Scalability in Reinforcement Learning for Quadruped Vision-Language-Action Models
by: Zhao, Han, et al.
Published: (2025)

Do What You Say: Steering Vision-Language-Action Models via Runtime Reasoning-Action Alignment Verification
by: Wu, Yilin, et al.
Published: (2025)

Action Hallucination in Generative Vision-Language-Action Models
by: Soh, Harold, et al.
Published: (2026)

Align-Then-stEer: Adapting the Vision-Language Action Models through Unified Latent Guidance
by: Zhang, Yang, et al.
Published: (2025)

The Impact of Language Mixing on Bilingual LLM Reasoning
by: Li, Yihao, et al.
Published: (2025)

Continually Evolving Skill Knowledge in Vision Language Action Model
by: Wu, Yuxuan, et al.
Published: (2025)

Adaptive Capacity Allocation for Vision Language Action Fine-tuning
by: Kim, Donghoon, et al.
Published: (2026)

AnoleVLA: Lightweight Vision-Language-Action Model with Deep State Space Models for Mobile Manipulation
by: Takagi, Yusuke, et al.
Published: (2026)

Verifier-free Test-Time Sampling for Vision Language Action Models
by: Jang, Suhyeok, et al.
Published: (2025)

Exploring the Adversarial Vulnerabilities of Vision-Language-Action Models in Robotics
by: Wang, Taowen, et al.
Published: (2024)

Probing a Vision-Language-Action Model for Symbolic States and Integration into a Cognitive Architecture
by: Lu, Hong, et al.
Published: (2025)

AR-VLA: True Autoregressive Action Expert for Vision-Language-Action Models
by: Hu, Yutong, et al.
Published: (2026)

ActionFlow: A Pipelined Action Acceleration for Vision Language Models on Edge
by: Dai, Yuntao, et al.
Published: (2025)

ALOE: Action-Level Off-Policy Evaluation for Vision-Language-Action Model Post-Training
by: Yang, Rushuai, et al.
Published: (2026)

Adversarial Attacks on Robotic Vision Language Action Models
by: Jones, Eliot Krzysztof, et al.
Published: (2025)

Survey of Vision-Language-Action Models for Embodied Manipulation
by: Li, Haoran, et al.
Published: (2025)

Mean-Flow based One-Step Vision-Language-Action
by: Chen, Yang, et al.
Published: (2026)

Developing Vision-Language-Action Model from Egocentric Videos
by: Yoshida, Tomoya, et al.
Published: (2025)

Emergence of Human to Robot Transfer in Vision-Language-Action Models
by: Kareer, Simar, et al.
Published: (2025)

SAFE: Multitask Failure Detection for Vision-Language-Action Models
by: Gu, Qiao, et al.
Published: (2025)

villa-X: Enhancing Latent Action Modeling in Vision-Language-Action Models
by: Chen, Xiaoyu, et al.
Published: (2025)

WMPO: World Model-based Policy Optimization for Vision-Language-Action Models
by: Zhu, Fangqi, et al.
Published: (2025)

V-VLAPS: Value-Guided Planning for Vision-Language-Action Models
by: Ren, Ke, et al.
Published: (2026)

Experiences from Benchmarking Vision-Language-Action Models for Robotic Manipulation
by: Zhang, Yihao, et al.
Published: (2025)

Hierarchical Vision Language Action Model Using Success and Failure Demonstrations
by: Park, Jeongeun, et al.
Published: (2025)

Pure Vision Language Action (VLA) Models: A Comprehensive Survey
by: Zhang, Dapeng, et al.
Published: (2025)

Do What? Teaching Vision-Language-Action Models to Reject the Impossible
by: Hsieh, Wen-Han, et al.
Published: (2025)

Towards Backdoor-Based Ownership Verification for Vision-Language-Action Models
by: Sun, Ming, et al.
Published: (2026)

Expertise need not monopolize: Action-Specialized Mixture of Experts for Vision-Language-Action Learning
by: Shen, Weijie, et al.
Published: (2025)

FedVLA: Federated Vision-Language-Action Learning with Dual Gating Mixture-of-Experts for Robotic Manipulation
by: Miao, Cui, et al.
Published: (2025)

Action-aware Dynamic Pruning for Efficient Vision-Language-Action Manipulation
by: Pei, Xiaohuan, et al.
Published: (2025)

KineVLA: Towards Kinematics-Aware Vision-Language-Action Models with Bi-Level Action Decomposition
by: Han, Gaoge, et al.
Published: (2026)

SafeVLA: Towards Safety Alignment of Vision-Language-Action Model via Constrained Learning
by: Zhang, Borong, et al.
Published: (2025)

LoHoVLA: A Unified Vision-Language-Action Model for Long-Horizon Embodied Tasks
by: Yang, Yi, et al.
Published: (2025)

RePO-VLA: Recovery-Driven Policy Optimization for Vision-Language-Action Models
by: Liufu, Weijia, et al.
Published: (2026)