Di Liu

7

Papers

90

Total Citations

Papers (7)

RetrievalAttention: Accelerating Long-Context LLM Inference via Vector Retrieval

LUCAS: Layered Universal Codec Avatars

Instantaneous Perception of Moving Objects in 3D

Temporal Action Localization with Cross Layer Task Decoupling and Refinement

T2Bs: Text-to-Character Blendshapes via Video Generation

Show and Segment: Universal Medical Image Segmentation via In-Context Learning

Layout-Agnostic Scene Text Image Synthesis with Diffusion Models