aiqbench analysis
This research introduces CSES, a training-free method for Large Vision-Language Models (LVLMs) that adaptively selects keyframes to reduce computational overhead during long-video analysis. By formulating frame selection as a coverage problem with monotone and submodular objectives, the approach significantly cuts inference costs while maintaining accuracy across multiple benchmarks. The technique offers immediate practical value by accelerating processing speeds without requiring model retraining or sacrificing performance on downstream tasks.
Coverage-Driven Adaptive Keyframe Selection for Video Understanding
arXiv:2608.00714v1 Announce Type: cross Abstract: Recent advances in large vision-language models (LVLMs) have enabled long-video understanding and analysis. However, processing the large number of frames in a video incurs substantial computational overhead. Existing methods reduce LVLM inference costs by scoring frame-query relevance before inference and selecting keyframes accordingly. Nevertheless, the distribution of relevant frames varies across queries, and these methods often need to score hundreds or thousands of frames. To address this limitation, we propose CSES, a training-free semantic keyframe selector that adaptively determines the numbers of frames to score and keyframes to select. CSES estimates the prominence of the frame-query relevance profile to guide active acquisition and adapt the temporal coverage of each input. It then formulates keyframe selection as a coverage problem that jointly accounts for semantic relevance, temporal redundancy, and visual redundancy. Active acquisition and keyframe selection terminate based on coverage saturation. The selection objective is monotone and submodular, enabling greedy optimization with a standard approximation guarantee. Experiments with four LVLMs on two benchmarks show that our method preserves accuracy while scoring $4$-$13\times$ fewer frames and selecting $18.4\%$-$20.5\%$ fewer input keyframes than existing baselines. CSES further achieves a $3.1$-$5.4\times$ speedup in frame selection over baselines.
Abstract:Recent advances in large vision-language models (LVLMs) have enabled long-video understanding and analysis. However, processing the large number of frames in a video incurs substantial computational overhead. Existing methods reduce LVLM inference costs by scoring frame-query relevance before inference and selecting keyframes accordingly. Nevertheless, the distribution of relevant frames varies across queries, and these methods often need to score hundreds or thousands of frames. To address this limitation, we propose CSES, a training-free semantic keyframe selector that adaptively determines the numbers of frames to score and keyframes to select. CSES estimates the prominence of the frame-query relevance profile to guide active acquisition and adapt the temporal coverage of each input. It then formulates keyframe selection as a coverage problem that jointly accounts for semantic relevance, temporal redundancy, and visual redundancy. Active acquisition and keyframe selection terminate based on coverage saturation. The selection objective is monotone and submodular, enabling greedy optimization with a standard approximation guarantee. Experiments with four LVLMs on two benchmarks show that our method preserves accuracy while scoring $4$-$13\times$ fewer frames and selecting $18.4\%$-$20.5\%$ fewer input keyframes than existing baselines. CSES further achieves a $3.1$-$5.4\times$ speedup in frame selection over baselines.
Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
arXiv:2608.00714 [cs.CV]
(or arXiv:2608.00714v1 [cs.CV] for this version)
https://doi.org/10.48550/arXiv.2608.00714 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 293 words
Related coverage
Reasoning with Memory: A Temporal Granularity-Adaptive Framework for Training-Free Long Video Understanding
arXiv · Jul 29, 2026
VideoX-Qwen: Data-Centric Instruction-Based Video Editing
arXiv · Sep 24, 2026
[AINews] not much happened today
Latent Space · Sep 10, 2026
Skydio CEO Adam Bry on why Silicon Valley shouldn’t draw red lines for drone use
The Verge · Jun 15, 2026