OmniDTBC: Dynamic Token Budget Compression for Video-Audio Large Language Model
Open Access
- 9 November 2026
- conference paper
- Published by Association for Computing Machinery (ACM)
- p. 8760-8769
Abstract
No abstract availableFunding Information
- National Natural Science Foundation of China (62406068)
- Guangdong S&T Program (2026B0101100002)
- Shenzhen Science and Technology Program (SYSRD 20250529113401002)
- Shenzhen Science and Technology Program (CJGJZD 20250902091203005)
This publication has 6 references indexed in Scilit:
- RoBiFusion: A Robust and Bidirectional Interaction Camera-LiDAR 3D Object Detection Framework2025
- Audio-centric Video Understanding Benchmark without Text Shortcut2025
- LMMs-Eval: Reality Check on the Evaluation of Large Multimodal Models2025
- DEITalk: Speech-Driven 3D Facial Animation with Dynamic Emotional Intensity Modeling2024
- Few-Shot Video Classification via Representation Fusion and Promotion Learning2023
- Pengi: An Audio Language Model for Audio Tasks2023