Keda Tao

4 papers · 2025–2026 · 2 venues · across top CS/AI venues

Authors are matched by name, so researchers who share a name ("Yang Liu", "Wei Wang") appear as a single profile and their paper counts, venues and achievements are combined. How identity is handled.

Achievements

Jump to papers ↓

🌍 Conference Polyglot · Bronze 🌉 Interdisciplinary Bridge · Bronze 🗺️ Taxonomy Completionist (16)

Venues

CVPR (3) ICLR (1)

Top co-authors

Huan Wang (3) Kele Shao (2) Xueyi Chen (1) Xiucheng Wang (1) Jinjin Gu (1) Yang Sui (1) Jian Liu (1) Nan Cheng (1) Yulun Zhang (1) Haoxuan You (1)

Research topics

Artificial Intelligence (2) Computer Vision (2) Deep Learning (2) Machine Learning (2) Application Areas (2) Core AI (2) Models (2) Processing (2) Natural Language Processing (1) Analysis (1) Learning Types (1) Optimization & Theory (1) Resources & Methods (1)

Keywords

token compression (3) diffusion model (1) memory efficiency (1) vision-language model (1) inference efficiency (1) kv cache compression (1) inference acceleration (1) kv cache (1) identity preservation (1) face restoration (1) inference speedup (1) facial attribute (1) temporal compression (1) kv cache reduction (1) video large language model (1) dynamic pruning (1) temporal redundancy (1) cross-modal similarity (1) dynamic token (1) streaming video understanding (1)

Papers

OmniZip: Audio-Guided Dynamic Token Compression for Fast Omnimodal Large Language Models CVPR 2026 StreamingTOM: Streaming Token Compression for Efficient Video Understanding CVPR 2026 DyCoke: Dynamic Compression of Tokens for Fast Video Large Language Models CVPR 2025 Overcoming False Illusions in Real-World Face Restoration with Multi-Modal Guided Diffusion Model ICLR 2025