# Vision--Language Signals in Constrained RL: Safety Gains Without Anticipation

arXiv: https://arxiv.org/abs/2609.34041 · HTML: https://mlsanity.com/paper/2609.34041/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Placed in RL for autonomous driving and traffic, one of 15 directions in Deep reinforcement learning methods.
- Builds on 29 of its 40 references that are in the map.
- First posted on arXiv 28 Sep 2026.

## Placed in

- Deep reinforcement learning methods / [RL for autonomous driving and traffic](https://mlsanity.com/deeprl/rl-for-autonomous-driving-and-traffic/)

## Builds on (most cited first, 29 of 29)

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [Proximal Policy Optimization Algorithms](https://mlsanity.com/paper/1707.06347/) (2017)
- [Soft Actor-Critic: Off-Policy Maximum Entropy Deep Reinforcement Learning with a Stochastic Actor](https://mlsanity.com/paper/1801.01290/) (2018)
- [RT-2: Vision-Language-Action Models Transfer Web Knowledge to Robotic Control](https://mlsanity.com/paper/2307.15818/) (2023)
- [Exploration by Random Network Distillation](https://mlsanity.com/paper/1810.12894/) (2018)
- [Constrained Policy Optimization](https://mlsanity.com/paper/1705.10528/) (2017)
- [Deep Reinforcement Learning at the Edge of the Statistical Precipice](https://mlsanity.com/paper/2108.13264/) (2021)
- [Deep Reinforcement Learning that Matters](https://mlsanity.com/paper/1709.06560/) (2017)
- [VoxPoser: Composable 3D Value Maps for Robotic Manipulation with Language Models](https://mlsanity.com/paper/2307.05973/) (2023)
- [Eureka: Human-Level Reward Design via Coding Large Language Models](https://mlsanity.com/paper/2310.12931/) (2023)
- [MineDojo: Building Open-Ended Embodied Agents with Internet-Scale Knowledge](https://mlsanity.com/paper/2206.08853/) (2022)
- [Responsive Safety in Reinforcement Learning by PID Lagrangian Methods](https://mlsanity.com/paper/2007.03964/) (2020)
- [WinCLIP: Zero-/Few-Shot Anomaly Classification and Segmentation](https://mlsanity.com/paper/2303.14814/) (2023)
- [MetaDrive: Composing Diverse Driving Scenarios for Generalizable Reinforcement Learning](https://mlsanity.com/paper/2109.12674/) (2021)
- [Reward Design with Language Models](https://mlsanity.com/paper/2303.00001/) (2023)
- [Projection-Based Constrained Policy Optimization](https://mlsanity.com/paper/2010.03152/) (2020)
- [Vision-Language Models are Zero-Shot Reward Models for Reinforcement Learning](https://mlsanity.com/paper/2310.12921/) (2023)
- [Text2Reward: Reward Shaping with Language Models for Reinforcement Learning](https://mlsanity.com/paper/2309.11489/) (2023)
- [Safety-Gymnasium: A Unified Safe Reinforcement Learning Benchmark](https://mlsanity.com/paper/2310.12567/) (2023)
- [CRPO: A New Approach for Safe Reinforcement Learning with Convergence Guarantee](https://mlsanity.com/paper/2011.05869/) (2020)
- [OmniSafe: An Infrastructure for Accelerating Safe Reinforcement Learning Research](https://mlsanity.com/paper/2305.09304/) (2023)
- [SafeVLA: Towards Safety Alignment of Vision-Language-Action Model via Constrained Learning](https://mlsanity.com/paper/2503.03480/) (2025)
- [Saute RL: Almost Surely Safe Reinforcement Learning Using State Augmentation](https://mlsanity.com/paper/2202.06558/) (2022)
- [Penalized Proximal Policy Optimization for Safe Reinforcement Learning](https://mlsanity.com/paper/2205.11814/) (2022)
- [Constrained Update Projection Approach to Safe Policy Optimization](https://mlsanity.com/paper/2209.07089/) (2022)
- [VLM-RL: A Unified Vision Language Models and Reinforcement Learning Framework for Safe Autonomous Driving](https://mlsanity.com/paper/2412.15544/) (2024)
- [VLM-SAFE: Vision-Language Model-Guided Safety-Aware Reinforcement Learning with World Models for Autonomous Driving](https://mlsanity.com/paper/2505.16377/) (2025)
- [Updating Robot Safety Representations Online from Natural Language Feedback](https://mlsanity.com/paper/2409.14580/) (2024)
- [Dynamic Control Barrier Function Regulation with Vision-Language Models for Safe, Adaptive, and Realtime Visual Navigation](https://mlsanity.com/paper/2603.21142/) (2026)

## Built on by (most cited first, 1 of 1)

- [It Is Not Seeing the Hazard: A Frozen Vision-Language Safety Score Measures Its Caption Bank](https://mlsanity.com/paper/2610.09517/) (2026)

## Abstract (arXiv)

Safe reinforcement learning seeks policies that maximise task performance while satisfying safety constraints. In driving benchmarks, however, collision costs typically appear only at the time of collision, providing no advance warning of an approaching hazard. Frozen vision--language models can provide dense semantic feedback, yet it remains unclear whether their scores anticipate collisions and which component drives an observed safety improvement. Episodic cost can also favour policies that make little task progress. To address these gaps, we propose VLM-Safe-RL, a framework that integrates frozen CLIP signals into PPO-Lagrangian through reward shaping and an augmented multiplier update. On MetaDrive Hard, which combines the densest traffic with the largest map, the catastrophe rate falls from 31.6% to 19.4%. FormulaOne-L2 analysis finds no evidence that the CLIP signals anticipate collisions and shows that the VLM term has a negligible effect on the Lagrange multiplier. These findings show a conditional reduction in observed catastrophe rate without evidence of collision anticipation.