Picture for Xingjun Ma

Xingjun Ma

NAP-Tuning: Neural Augmented Prompt Tuning for Adversarially Robust Vision-Language Models

Add code
Jun 15, 2025
Viaarxiv icon

GenBreak: Red Teaming Text-to-Image Generators Using Large Language Models

Add code
Jun 11, 2025
Viaarxiv icon

RedRFT: A Light-Weight Benchmark for Reinforcement Fine-Tuning-Based Red Teaming

Add code
Jun 04, 2025
Viaarxiv icon

From Failures to Fixes: LLM-Driven Scenario Repair for Self-Evolving Autonomous Driving

Add code
May 28, 2025
Viaarxiv icon

JailBound: Jailbreaking Internal Safety Boundaries of Vision-Language Models

Add code
May 26, 2025
Viaarxiv icon

SAMA: Towards Multi-Turn Referential Grounded Video Chat with Large Language Models

Add code
May 24, 2025
Viaarxiv icon

SafeVid: Toward Safety Aligned Video Large Multimodal Models

Add code
May 17, 2025
Viaarxiv icon

X-Transfer Attacks: Towards Super Transferable Adversarial Attacks on CLIP

Add code
May 08, 2025
Viaarxiv icon

Toward Generalizable Evaluation in the LLM Era: A Survey Beyond Benchmarks

Add code
Apr 26, 2025
Viaarxiv icon

A Comprehensive Survey in LLM(-Agent) Full Stack Safety: Data, Training and Deployment

Add code
Apr 22, 2025
Viaarxiv icon