Seeing the Forest and the Trees: Query-Aware Tokenizer for Long-Video Multimodal Language Models

• URL: http://arxiv.org/abs/2511.11910v2
• Date: Fri, 21 Nov 2025 17:08:33 GMT
• Title: Seeing the Forest and the Trees: Query-Aware Tokenizer for Long-Video Multimodal Language Models
• Authors: Siyou Li, Huanan Wu, Juexi Shao, Yinghao Ma, Yujian Gan, Yihao Luo, Yuwei Wang, Dong Nie, Lu Wang, Wengqing Wu, Le Zhang, Massimo Poesio, Juntao Yu,
• Abstract summary: We present QTSplus, a visual token selection module for long video understanding scenarios.<n>It is integrated into Qwen2.5-VL and compresses the vision stream by up to textbf89% and reduces end-to-end latency by textbf28% on long videos.<n>Results show that QTSplus is an effective, general mechanism for scaling MLLMs to realworld long-video scenarios.
• Score: 24.875526594002434
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Despite the recent advances in the video understanding ability of multimodal large language models (MLLMs), long video understanding remains a challenge. One of the main issues is that the number of vision tokens grows linearly with video length, which causes an explosion in attention cost, memory, and latency. To solve this challenge, we present Query-aware Token Selector (\textbf{QTSplus}), a lightweight yet powerful visual token selection module that serves as an information gate between the vision encoder and LLMs. Given a text query and video tokens, QTSplus dynamically selects the most important visual evidence for the input text query by (i) scoring visual tokens via cross-attention, (ii) \emph{predicting} an instance-specific retention budget based on the complexity of the query, and (iii) \emph{selecting} Top-$n$ tokens with a differentiable straight-through estimator during training and a hard gate at inference. Furthermore, a small re-encoder preserves temporal order using absolute time information, enabling second-level localization while maintaining global coverage. Integrated into Qwen2.5-VL, QTSplus compresses the vision stream by up to \textbf{89\%} and reduces end-to-end latency by \textbf{28\%} on long videos. The evaluation on eight long video understanding benchmarks shows near-parity accuracy overall when compared with the original Qwen models and outperforms the original model by \textbf{+20.5} and \textbf{+5.6} points respectively on TempCompass direction and order accuracies. These results show that QTSplus is an effective, general mechanism for scaling MLLMs to real-world long-video scenarios while preserving task-relevant evidence.

Related papers

• TrajTok: Learning Trajectory Tokens enables better Video Understanding [63.1260672430712]
  Tokenization in video models, typically through patchification, generates an excessive and redundant number of tokens.<n>We propose TrajTok, an end-to-end video tokenizer module that is fully integrated and co-trained with video models for a downstream objective.<n>We show that it can be seamlessly integrated as either a probing head for pretrained visual features (TrajAdapter) or an alignment connector in vision-language models (TrajVLM) with especially strong performance in long-video reasoning.
  arXiv  Detail & Related papers  (2026-02-26T09:15:34Z)
• Universal Video Temporal Grounding with Generative Multi-modal Large Language Models [59.781211641591405]
  This paper presents a computational model for universal video temporal grounding, which accurately localizes temporal moments in videos based on natural language queries.<n>We propose UniTime, a robust and universal video grounding model leveraging the strong vision-language understanding capabilities of generative Multi-modal Large Language Models (MLLMs)<n>Our model effectively handles videos of diverse views, genres, and lengths while comprehending complex language queries.
  arXiv  Detail & Related papers  (2025-06-23T17:53:18Z)
• Sparse-to-Dense: A Free Lunch for Lossless Acceleration of Video Understanding in LLMs [25.13186579764434]
  We introduce Sparse-to-Dense (StD), a novel decoding strategy that integrates two distinct modules.<n>StD is a tuning-free, plug-and-play solution that achieves up to a 1.94$times$ walltime speedup in video processing.
  arXiv  Detail & Related papers  (2025-05-25T14:09:28Z)
• Multimodal Long Video Modeling Based on Temporal Dynamic Context [13.979661295432964]
  We propose a dynamic long video encoding method utilizing the temporal relationship between frames, named Temporal Dynamic Context (TDC)<n>We segment the video into semantically consistent scenes based on inter-frame similarities, then encode each frame into tokens using visual-audio encoders.<n>To handle extremely long videos, we propose a training-free chain-of-thought strategy that progressively extracts answers from multiple video segments.
  arXiv  Detail & Related papers  (2025-04-14T17:34:06Z)
• VQToken: Neural Discrete Token Representation Learning for Extreme Token Reduction in Video Large Language Models [35.38573641029626]
  We introduce the novel task of Extreme Short Token Reduction, which aims to represent entire videos using a minimal set of discrete tokens.<n>On the Extreme Short Token Reduction task, our VQToken compresses sequences to just 0.07 percent of their original length while incurring only a 0.66 percent drop in accuracy on the NextQA-MC benchmark.
  arXiv  Detail & Related papers  (2025-03-21T09:46:31Z)
• BIMBA: Selective-Scan Compression for Long-Range Video Question Answering [46.199493246921435]
  Video Question Answering (VQA) in long videos poses the key challenge of extracting relevant information.<n>We introduce BIMBA, an efficient state-space model to handle long-form videos.
  arXiv  Detail & Related papers  (2025-03-12T17:57:32Z)
• QuoTA: Query-oriented Token Assignment via CoT Query Decouple for Long Video Comprehension [86.0749609778104]
  We propose QuoTA, an ante-hoc training-free modular that extends existing large video-language models.<n>QuoTA strategically allocates frame-level importance scores based on query relevance.<n>We decouple the query through Chain-of-Thoughts reasoning to facilitate more precise LVLM-based frame importance scoring.
  arXiv  Detail & Related papers  (2025-03-11T17:59:57Z)
• HierarQ: Task-Aware Hierarchical Q-Former for Enhanced Video Understanding [14.464718780172582]
  We introduce HierarQ, a task-aware hierarchical Q-Former based framework that sequentially processes frames to bypass the need for frame sampling.<n>We introduce a lightweight two-stream language-guided feature modulator to incorporate task awareness in video understanding.<n>Extensive evaluations on 10 video benchmarks across video understanding, question answering, and captioning tasks demonstrate HierarQ's state-of-the-art performance.
  arXiv  Detail & Related papers  (2025-03-11T16:21:23Z)
• STORM: Token-Efficient Long Video Understanding for Multimodal LLMs [116.4479155699528]
  STORM is a novel architecture incorporating a dedicated temporal encoder between the image encoder and the Video-LLMs.<n>We show that STORM achieves state-of-the-art results across various long video understanding benchmarks.
  arXiv  Detail & Related papers  (2025-03-06T06:17:38Z)
• VideoLLM-MoD: Efficient Video-Language Streaming with Mixture-of-Depths Vision Computation [66.00245701441547]
  We introduce a novel approach to reduce vision compute by leveraging redundant vision tokens "skipping layers" rather than decreasing the number of vision tokens. Our method, VideoLLM-MoD, is inspired by mixture-of-depths LLMs and addresses the challenge of numerous vision tokens in long-term or streaming video.
  arXiv  Detail & Related papers  (2024-08-29T17:21:58Z)
• Long Context Transfer from Language to Vision [74.78422371545716]
  Video sequences offer valuable temporal information, but existing large multimodal models (LMMs) fall short in understanding extremely long videos. In this paper, we approach this problem from the perspective of the language model. By simply extrapolating the context length of the language backbone, we enable LMMs to comprehend orders of magnitude more visual tokens without any video training.
  arXiv  Detail & Related papers  (2024-06-24T17:58:06Z)
• LongVLM: Efficient Long Video Understanding via Large Language Models [55.813206751150716]
  LongVLM is a simple yet powerful VideoLLM for long video understanding. We encode video representations that incorporate both local and global information. Our model produces more precise responses for long video understanding.
  arXiv  Detail & Related papers  (2024-04-04T11:33:29Z)
• Long Video Understanding with Learnable Retrieval in Video-Language Models [48.3525267216256]
  We introduce a learnable retrieval-based video-language model (R-VLM) for efficient long video understanding.<n>Specifically, given a question (Query) and a long video, our model identifies and selects the most relevant K video chunks.<n>This effectively reduces the number of video tokens, eliminates noise interference, and enhances system performance.
  arXiv  Detail & Related papers  (2023-12-08T09:48:36Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.