Question-in-the-Loop: Interleaved Query Tokens as Visual Summarizers for Video LLMs
Open Access
- 9 November 2026
- conference paper
- Published by Association for Computing Machinery (ACM)
- p. 4077-4086
Abstract
No abstract availableFunding Information
- National Natural Science Foundation of China (62502256)
- National Natural Science Foundation of China (62320106007)
- National Natural Science Foundation of China (62576216)
- Guangdong Provincial Key Laboratory (2023B1212060076)
This publication has 21 references indexed in Scilit:
- TokenPacker: Efficient Visual Projector for Multimodal LLMInternational Journal of Computer Vision, 2025
- VideoTree: Adaptive Tree-based Video Representation for LLM Reasoning on Long Videos2025
- DyCoke : Dynamic Compression of Tokens for Fast Video Large Language Models2025
- DivPrune: Diversity-based Visual Token Pruning for Large Multimodal Models2025
- MLVU: Benchmarking Multi-task Long Video Understanding2025
- VisionZip: Longer is Better but Not Necessary in Vision Language Models2025
- Hybrid-Level Instruction Injection for Video Token Compression in Multi-modal Large Language Models2025
- An Image is Worth 1/2 Tokens After Layer 2: Plug-and-Play Inference Acceleration for Large Vision-Language Models2024
- MVBench: A Comprehensive Multi-modal Video Understanding Benchmark2024
- MovieChat: From Dense Token to Sparse Memory for Long Video Understanding2024