Hugging Face daily papers·18d agoWhy Is Video Still So Expensive? A Survey of Inference-Efficiency Mechanisms in Video and Audiovisual LLMs#frame-sampling#inference-efficiency#multimodal
arXiv cs.CR·19d agoDo Input-Level Defenses Transfer to Observation-Level Attacks on VideoLLMs?#adversarial-robustness#defenses#deftevalAI safety & security
