Video-XL-2: Towards Very Long-Video Understanding Through Task-Aware KV Sparsification

• URL: http://arxiv.org/abs/2506.19225v1
• Date: Tue, 24 Jun 2025 01:19:56 GMT
• Title: Video-XL-2: Towards Very Long-Video Understanding Through Task-Aware KV Sparsification
• Authors: Minghao Qin, Xiangrui Liu, Zhengyang Liang, Yan Shu, Huaying Yuan, Juenjie Zhou, Shitao Xiao, Bo Zhao, Zheng Liu,
• Abstract summary: Video-XL-2 is a novel MLLM that delivers superior cost-effectiveness for long-video understanding based on task-aware KV sparsification.<n>It is capable of processing over 10,000 frames on a single NVIDIA A100 (80GB) GPU and thousands of frames in just a few seconds.
• Score: 9.615466029246694
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Multi-modal large language models (MLLMs) models have made significant progress in video understanding over the past few years. However, processing long video inputs remains a major challenge due to high memory and computational costs. This makes it difficult for current models to achieve both strong performance and high efficiency in long video understanding. To address this challenge, we propose Video-XL-2, a novel MLLM that delivers superior cost-effectiveness for long-video understanding based on task-aware KV sparsification. The proposed framework operates with two key steps: chunk-based pre-filling and bi-level key-value decoding. Chunk-based pre-filling divides the visual token sequence into chunks, applying full attention within each chunk and sparse attention across chunks. This significantly reduces computational and memory overhead. During decoding, bi-level key-value decoding selectively reloads either dense or sparse key-values for each chunk based on its relevance to the task. This approach further improves memory efficiency and enhances the model's ability to capture fine-grained information. Video-XL-2 achieves state-of-the-art performance on various long video understanding benchmarks, outperforming existing open-source lightweight models. It also demonstrates exceptional efficiency, capable of processing over 10,000 frames on a single NVIDIA A100 (80GB) GPU and thousands of frames in just a few seconds.

Related papers

• Token-Efficient Long Video Understanding for Multimodal LLMs [101.70681093383365]
  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)
• VideoChat-Flash: Hierarchical Compression for Long-Context Video Modeling [43.485687038460895]
  Long-context video modeling is critical for multimodal large language models (MLLMs)<n>This paper aims to address this issue from aspects of model architecture, training data, training strategy and evaluation benchmark.<n>We build a powerful video MLLM named VideoChat-Flash, which shows a leading performance on both mainstream long and short video benchmarks.
  arXiv  Detail & Related papers  (2024-12-31T18:01:23Z)
• ReTaKe: Reducing Temporal and Knowledge Redundancy for Long Video Understanding [55.320254859515714]
  ReTaKe enables VideoLLMs to process 8 times longer frames (up to 2048), similar-sized models by 3-5% and even rivaling much larger ones on VideoMME, MLVU, LongVideoBench, and LVBench.<n>Our code is available at https://github.com/SCZwangxiao/video-ReTaKe.
  arXiv  Detail & Related papers  (2024-12-29T15:42:24Z)
• Look Every Frame All at Once: Video-Ma$^2$mba for Efficient Long-form Video Understanding with Multi-Axis Gradient Checkpointing [52.050036778325094]
  Video-Ma$2$mba is a novel architecture that incorporates State Space Models (SSMs) within the Mamba-2 framework.<n>Our approach significantly reduces the memory footprint compared to standard gradient checkpointing.<n>By maintaining a detailed capture of temporal dynamics, our model improves the accuracy and relevance of responses in long video understanding tasks.
  arXiv  Detail & Related papers  (2024-11-29T04:12:13Z)
• Video-XL: Extra-Long Vision Language Model for Hour-Scale Video Understanding [25.61734041983714]
  Video-XL is a novel approach that leverages MLLMs' inherent key-value sparsification capacity to condense the visual input.<n>Video-XL's effectiveness is verified from three aspects. First, it achieves a superior long-video understanding capability, outperforming state-of-the-art models of comparable sizes.
  arXiv  Detail & Related papers  (2024-09-22T15:13:31Z)
• VideoLLaMB: Long-context Video Understanding with Recurrent Memory Bridges [42.555895949250704]
  VideoLLaMB is a novel framework that utilizes temporal memory tokens within bridge layers to allow for the encoding of entire video sequences. SceneTilling algorithm segments videos into independent semantic units to preserve semantic integrity. In terms of efficiency, VideoLLaMB, trained on 16 frames, supports up to 320 frames on a single Nvidia A100 GPU.
  arXiv  Detail & Related papers  (2024-09-02T08:52:58Z)
• A Simple Recipe for Contrastively Pre-training Video-First Encoders Beyond 16 Frames [57.758863967770594]
  We build on the common paradigm of transferring large-scale, image--text models to video via shallow temporal fusion.<n>We expose two limitations to the approach: (1) decreased spatial capabilities, likely due to poor video--language alignment in standard video datasets, and (2) higher memory consumption, bottlenecking the number of frames that can be processed.
  arXiv  Detail & Related papers  (2023-12-12T16:10:19Z)
• EVEREST: Efficient Masked Video Autoencoder by Removing Redundant Spatiotemporal Tokens [57.354304637367555]
  We present EVEREST, a surprisingly efficient MVA approach for video representation learning. It finds tokens containing rich motion features and discards uninformative ones during both pre-training and fine-tuning. Our method significantly reduces the computation and memory requirements of MVA.
  arXiv  Detail & Related papers  (2022-11-19T09:57:01Z)
• Deep Unsupervised Key Frame Extraction for Efficient Video Classification [63.25852915237032]
  This work presents an unsupervised method to retrieve the key frames, which combines Convolutional Neural Network (CNN) and Temporal Segment Density Peaks Clustering (TSDPC) The proposed TSDPC is a generic and powerful framework and it has two advantages compared with previous works, one is that it can calculate the number of key frames automatically. Furthermore, a Long Short-Term Memory network (LSTM) is added on the top of the CNN to further elevate the performance of classification.
  arXiv  Detail & Related papers  (2022-11-12T20:45:35Z)

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.