Safety as a Budget: Calibrating the Control Channel of Constrained GRPO for Vision-Language Models
cs.AI, cs.CV
Submitted: 2026-11-12
Updated: 2026-11-12
Terminology
Sources
- Constrained Group Relative Policy Optimization
- Just Ask Jev: Reinforcement Learning for Calibrated Decisions as a Zero-Shot Detector of AI Alignment Failures
- Llama Guard: LLM-based Input-Output Safeguard for Human-AI Conversations
- CARGO-VL: Counterfactual Arbitration with Risk-Constrained Group Optimization for Vision-Language Models
- DecisionRobust: Benchmarking Robustness for JEV Style Decision Models Across Text and Vision
- Visual Jev Rewards: Reference-Bound Verification for Multi-Subject Image Generation
- Same Attack, Different Judge: A First Controlled Adversarial Comparison of Jev, LLM Judges, and Purpose-Built Safety Classifiers
- LLM-as-Jev: LLMs Are Already Jev-Style Decision Models -- When and How to Fine-Tune Them
- Understanding R1-Zero-Like Training: A Critical Perspective
- JailBreakV: A Benchmark for Assessing the Robustness of MultiModal Large Language Models against Jailbreak Attacks
- DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models
- Hidden Risks of Jev: An Empirical Study of Security, Privacy, and Dual Use
Related papers
- Multi-Objective Structured Pruning of LLMs for Latency and Model Size Optimization
- MAVEN-T: Reinforced Heterogeneous Distillation for Real-Time Multi-Agent Trajectory Prediction
- Model Discovery Agent: LLM-assisted Bayesian experiment design for data-efficient discovery of mechanistic world models
- Embeddings for Preferences, Not Semantics
- Can We Trust Item Response Theory for AI Evaluation?
- Diagnosing Task Insensitivity in Language Agents