ShieldVLA is proposed, a safety-aligned fine-tuning framework for VLA models based on Hamilton-Jacobi (HJ) reachability that reduces cumulative safety cost by 57% on average and improves task success rate by +0.13 over SafeVLA.
Abstract
Vision-Language-Action (VLA) models demonstrate strong generalization in robotic manipulation and navigation, but existing fine-tuning methods provide limited safety guarantees. Current approaches primarily rely on Lagrangian optimization that enforces safety through soft penalties on expected cumulative cost, often resulting in residual constraint violations or overly conservative behavior. Moreover, learning safety in visual domains is challenging due to the absence of dense per-step safety annotations. We propose ShieldVLA, a safety-aligned fine-tuning framework for VLA models based on Hamilton-Jacobi (HJ) reachability. ShieldVLA learns a model-free approximation of the HJ reachability value function directly from visual observations to estimate the safe operating region. The learned safety critic gates policy optimization by separating reward maximization within feasible regions from recovery near unsafe states, avoiding persistent reward-cost trade-offs. To enable scalable supervision in visual environments, we introduce rubric-based VLM safety scores that convert semantic safety feedback into structured critic targets without requiring manual cost labels. Across five navigation and manipulation benchmarks spanning multiple VLA backbones, ShieldVLA reduces cumulative safety cost by 57% on average and improves task success rate by +0.13 over SafeVLA.
This work presents Vision-Language-Poisson-Safe Actions (VLPSA), a safety filtering framework that provides full-body safety for VLA policies in cluttered and dynamic environments without retraining, and deploys VLPSA on a Franka FR3 in cluttered scenes with dynamic obstacles and human interference.
Meg Wilkinson, E. Fourney, Joel W. Burdick et al.· 0 citations
Vision-Language-Action (VLA) models for autonomous driving rely heavily on successful expert demonstrations, leaving model-specific failures underexploited. Learning from these failures is hindered by unreliable diagnoses, poorly matched correction targets, and coarse rewards. We propose RefineDrive, a failure-guided p...
Zhe Sun, Zi-Yin Luo, Ye-Hao Lu et al.· 0 citations
ActSafeGuard is introduced, a differentiable and training-aligned safeguard layer for flow-matching based policies that integrates hard action feasibility into policy learning, not merely treating safety as an inference-time external component.
Jian-Ming Ma, Rong-Jun Jin, Xia-Xi Si et al.· 0 citations
VLM-Safe-RL is proposed, a framework that integrates frozen CLIP signals into PPO-Lagrangian through reward shaping and an augmented multiplier update, and shows a conditional reduction in observed catastrophe rate without evidence of collision anticipation.
Samuel T. K. Tetteh, Cody H. Fleming· arXiv.org· 1 citation
SafeBranch is proposed, a framework that aligns an embodied actor on safety through branch pairs constructed from the actor's own unsafe rollouts via environment rollback, achieving roughly ten times more safe successes than the untrained baseline on the unseen-object variant.
Hyunse Lee, Jiwoo Jeong, Haneul Lee et al.· 0 citations
Alignment-Guided Flow Transformer (AGFT) is presented, a novel framework that explicitly enforces tri-modal alignment through a dedicated alignment loss, bridging the representational gap across modalities and enhancing task adaptation.
Sheng-Chao Hu, Peng Wang, Qi-Yang Zhou et al.· 0 citations
Related blog posts
MIT News · Artificial Intelligence· news.mit.eduJul 30, 2026
From feet to fingertips — we are teaching robots intelligent whole-body control, fine dexterity, and teamwork to complete a broad range of complex tasks.