Enhancing Visual Token Representations for Video Large Language Models via Training-Free Spatial-Temporal Pooling and Gridding

• URL: http://arxiv.org/abs/2605.22078v1
• Date: Thu, 21 May 2026 07:16:45 GMT
• Title: Enhancing Visual Token Representations for Video Large Language Models via Training-Free Spatial-Temporal Pooling and Gridding
• Abstract summary: We propose ST-Grid, a training-Pool visual token enhancement method designed specifically for Video Large Language Models.<n>Our approach integrates Pyramid Temporal Gridding (PTG), which captures multi-temporal interactions through hierarchical temporal gridding.<n>Our method offers an efficient and plug-and-play solution for improving visual token representations.
• Score: 11.95854121762109
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Recent advances in Multimodal Large Language Models (MLLMs) have significantly advanced video understanding tasks, yet challenges remain in efficiently compressing visual tokens while preserving spatiotemporal interactions. Existing methods, such as LLaVA family, utilize simplistic pooling or interpolation techniques that overlook the intricate dynamics of visual tokens. To bridge this gap, we propose ST-GridPool, a novel training-free visual token enhancement method designed specifically for Video LLMs. Our approach integrates Pyramid Temporal Gridding (PTG), which captures multi-grained spatiotemporal interactions through hierarchical temporal gridding, and Norm-based Spatial Pooling (NSP), which preserves high-information visual regions by leveraging the correlation between token norms and semantic richness. Extensive experiments on various benchmarks demonstrate that ST-GridPool consistently enhances performance of Video LLMs without requiring costly retraining. Our method offers an efficient and plug-and-play solution for improving visual token representations. Our code is available in https://github.com/bingjunluo/ST-GridPool.

Related papers

• CoViPAL: Layer-wise Contextualized Visual Token Pruning for Large Vision-Language Models [75.88232735646018]
  Large Vision-Language Models (LVLMs) process multimodal inputs consisting of text tokens and vision tokens extracted from images or videos.<n>Existing methods attempt to prune redundant vision tokens, revealing substantial redundancy in visual representations.<n>We propose CoViPAL, a layer-wise contextualized visual token pruning method that employs a Plug-and-Play Pruning Module (PPM) to predict and remove redundant vision tokens before they are processed by the LVLM.
  arXiv  Detail & Related papers  (2025-08-24T07:47:00Z)
• TSPO: Temporal Sampling Policy Optimization for Long-form Video Language Understanding [25.675553077419274]
  Multimodal Language Models (MLLMs) have demonstrated significant progress in vision tasks, yet they still face challenges when processing long-duration inputs.<n>We propose Temporal Sampling Policy Optimization (TSPO), advancing MLLMs' long-form video-language understanding via reinforcement learning.<n>Our TSPO state-of-the-art performance across multiple long video understanding benchmarks, and shows transferable ability across different cutting-edge Video-MLLMs.
  arXiv  Detail & Related papers  (2025-08-06T12:03:36Z)
• Reinforcement Learning Tuning for VideoLLMs: Reward Design and Data Efficiency [56.475612147721264]
  We propose a dual-reward formulation that supervises both semantic and temporal reasoning through discrete and continuous reward signals.<n>We evaluate our approach across eight representative video understanding tasks, including VideoQA, Temporal Video Grounding, and Grounded VideoQA.<n>Results underscore the importance of reward design and data selection in advancing reasoning-centric video understanding with MLLMs.
  arXiv  Detail & Related papers  (2025-06-02T17:28:26Z)
• 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)
• AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning [19.68349294206012]
  Large language models (LLMs) have enabled the creation of multi-modal LLMs that exhibit strong comprehension of visual data such as images and videos.<n>We propose a training-free adaptive inference method for multi-modal LLMs that can accommodate a broad range of efficiency requirements.
  arXiv  Detail & Related papers  (2024-12-04T11:47:57Z)
• Free Video-LLM: Prompt-guided Visual Perception for Efficient Training-free Video LLMs [56.040198387038025]
  We present a novel prompt-guided visual perception framework (abbreviated as Free Video-LLM) for efficient inference of training-free video LLMs. Our method effectively reduces the number of visual tokens while maintaining high performance across multiple video question-answering benchmarks.
  arXiv  Detail & Related papers  (2024-10-14T12:35:12Z)
• ST-LLM: Large Language Models Are Effective Temporal Learners [58.79456373423189]
  Large Language Models (LLMs) have showcased impressive capabilities in text comprehension and generation. How to effectively encode and understand videos in video-based dialogue systems remains to be solved. We propose ST-LLM, an effective video-LLM baseline with spatial-temporal sequence modeling inside LLM.
  arXiv  Detail & Related papers  (2024-03-30T10:11:26Z)
• Efficient Temporal Extrapolation of Multimodal Large Language Models with Temporal Grounding Bridge [47.750073410717604]
  We introduce Temporal Grounding Bridge (TGB), a novel framework that bootstraps MLLMs with advanced temporal grounding capabilities. We validate TGB across seven video benchmarks and demonstrate substantial performance improvements compared with prior MLLMs. Our model, initially trained on sequences of four frames, effectively handles sequences up to 16 longer without sacrificing performance.
  arXiv  Detail & Related papers  (2024-02-25T10:27:46Z)
• Video-LaVIT: Unified Video-Language Pre-training with Decoupled Visual-Motional Tokenization [52.63845811751936]
  Video pre-training is challenging due to the modeling of its dynamics video. In this paper, we address such limitations in video pre-training with an efficient video decomposition. Our framework is both capable of comprehending and generating image and video content, as demonstrated by its performance across 13 multimodal benchmarks.
  arXiv  Detail & Related papers  (2024-02-05T16:30:49Z)

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.