2 papers
cs.LG2025
Internal Activation Revision: Safeguarding Vision Language Models Without Parameter Update
Qing Li, Jiahui Geng, Zongxiong Chen +3
Vision-language models (VLMs) demonstrate strong multimodal capabilities but have been found to be more susceptible to generating harmful content compared to their backbone large l…
cs.CL2024
Reference-free Hallucination Detection for Large Vision-Language Models
Qing Li, Jiahui Geng, Chenyang Lyu +3
Large vision-language models (LVLMs) have made significant progress in recent years. While LVLMs exhibit excellent ability in language understanding, question answering, and conver…