Papers for

content moderation services

Papers whose findings have a practical use for this group, as judged from the abstract. Open a paper to read what it means in practice.

MarKey improves keyframe selection for better long video understanding

MarKey: Marginal Utility Guided Greedy Keyframe Selection for Long Video Understanding

Abstract: Long-video understanding remains challenging for multimodal large language models (MLLMs) because densely encoding long frame sequences is computationally expensive, while uniform sampling under a limited visual budget can miss sparse yet decisive evidence. Recent training-free keyframe selection methods have enabled more efficient inference and yielded promising performance gains. However, many existing methods score frames largely in isolation without explicitly considering how each candidate complements the currently selected subset, potentially resulting in redundant selections and incomplete evidence coverage. To address this limitation, we propose MarKey, a training-free framework that formulates keyframe selection as subset-aware greedy optimization. At each iteration, MarKey scores each candidate using a tractable surrogate that jointly accounts for query relevance, marginal coverage gain, and context-dependent redundancy, and selects the frame with the highest utility. To make this iterative subset-aware evaluation efficient, MarKey uses a compact set of representative anchors to approximate full-video coverage and a bounded window of previously selected frames to limit context-dependent comparisons. Experiments on six benchmarks spanning holistic video understanding, human-centric video understanding, and open-ended video understanding demonstrate that MarKey consistently outperforms existing methods. Further analyses show robust gains across different MLLM backbones, model scales, and frame budgets.

Mon 14 SeptComputer Vision and Pattern RecognitionComputation and Language
The gist
Understanding very long videos is hard because analyzing every frame takes too much computing power, and just picking frames evenly can miss important moments. The authors present MarKey, a method that selects key frames by considering which new frame adds the most new information without repeating what’s already chosen. MarKey looks at how each frame helps cover the video’s content while avoiding redundancy, and does this efficiently with smart shortcuts. Tests show MarKey picks better frames than other methods for various video understanding tasks.
Open 2609.15408v1