Grammar-Constrained Decoding Can Jailbreak LLMs into Generating Malicious Code
Paper • 2606.11817 • Published • 18
Papers of Yitong
-
-
Lookahead-then-Verify: Reliable Constrained Decoding for Diffusion LLMs under Context-Free Grammars
Paper • 2602.00612 • Published • 1 -
Environmental Injection Attacks against GUI Agents in Realistic Dynamic Environments
Paper • 2509.11250 • Published -
DAVSP: Safety Alignment for Large Vision-Language Models via Deep Aligned Visual Safety Prompt
Paper • 2506.09353 • Published -
To See is Not to Master: Teaching LLMs to Use Private Libraries for Code Generation
Paper • 2603.15159 • Published -
Beyond Autoregression: An Empirical Study of Diffusion Large Language Models for Code Generation
Paper • 2509.11252 • Published • 1