Search papers, labs, and topics across Lattice.
Zhejiang University
5
0
7
VideoLLMs can improve their grasp of physical laws, but they still falter when faced with complex causal contexts.
Privacy-sensitive visual and textual information can be recovered from deep-layer LVLM hidden states, exposing significant risks in collaborative inference.
Persona conditioning in LLMs can be exploited to amplify inference costs by over 200 times, revealing a critical vulnerability in their deployment.
Achieving competitive performance in Vision-Language Models while reducing computational overhead could revolutionize their deployment on edge devices.
Stop retraining watermarks for every LoRA: LoRA-Key lets you inject a reusable, user-specific copyright key into text-to-image models without sacrificing image quality or style.
