Disagreement-Guided Agentic Data Curation for Multimodal Safety
cs.AI, cs.MA
Submitted: 2026-07-15
Updated: 2026-11-18
Terminology
Sources
- Security in LLM-as-a-Judge: A Comprehensive SoK
- Reading Between the Pixels: Linking Text-Image Embedding Alignment to Typographic Attack Success on Vision-Language Models
- Rethinking Fine-Tuning when Scaling Test-Time Compute: Limiting Confidence Improves Mathematical Reasoning
- Every Picture Tells a Dangerous Story: Memory-Augmented Multi-Agent Jailbreak Attacks on VLMs
- The Side Effects of Being Smart: Safety Risks in MLLMs' Multi-Image Reasoning
- Evolve the Method, Not the Prompts: Evolutionary Synthesis of Jailbreak Attacks on LLMs
- Jailbreaking LLMs & VLMs: Mechanisms, Evaluation, and Unified Defense
- Courtroom-Style Multi-Agent Debate with Progressive RAG and Role-Switching for Controversial Claim Verification
- ContextNav: Towards Agentic Multimodal In-Context Learning
- DetPO: In-Context Learning with Multi-Modal LLMs for Few-Shot Object Detection
- Aetheria: A multimodal interpretable content safety framework based on multi-agent debate and collaboration
- LlavaGuard: An Open VLM-based Framework for Safeguarding Vision Datasets and Models
- A Judge-free LLM Open-ended Generation Benchmark Based on the Distributional Hypothesis
- AI safety via debate
- Adversarial Attacks on Multimodal Large Language Models: A Comprehensive Survey
- Curriculum Guided Massive Multi Agent System Solving For Robust Long Horizon Tasks
- Evaluating Nova 2.0 Lite model under Amazon's Frontier Model Safety Framework
- BiasScope: Towards Automated Detection of Bias in LLM-as-a-Judge Evaluation
- T-MAP: Red-Teaming LLM Agents with Trajectory-aware Evolutionary Search
- ThinkSafe: Self-Generated Safety Alignment for Reasoning Models
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