# An LMM for Efficient Video Understanding via Reinforced Compression of Video Cubes

arXiv: https://arxiv.org/abs/2504.15270 · HTML: https://mlsanity.com/paper/2504.15270/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- Placed in Video understanding with MLLMs, one of 16 directions in Multimodal LLMs / vision-language models.
- Builds on 42 of its 54 references that are in the map.
- First posted on arXiv 21 Apr 2025.

## Placed in

- Multimodal LLMs / vision-language models / [Video understanding with MLLMs](https://mlsanity.com/mllm/video-understanding-with-mllms/)

## Builds on (most cited first, 30 of 42)

- [LLaMA: Open and Efficient Foundation Language Models](https://mlsanity.com/paper/2302.13971/) (2023)
- [Layer Normalization](https://mlsanity.com/paper/1607.06450/) (2016)
- [Qwen2.5-VL Technical Report](https://mlsanity.com/paper/2502.13923/) (2025)
- [Qwen2.5 Technical Report](https://mlsanity.com/paper/2412.15115/) (2024)
- [Sigmoid Loss for Language Image Pre-Training](https://mlsanity.com/paper/2303.15343/) (2023)
- [Fast R-CNN](https://mlsanity.com/paper/1504.08083/) (2015)
- [Categorical Reparameterization with Gumbel-Softmax](https://mlsanity.com/paper/1611.01144/) (2016)
- [LLaVA-OneVision: Easy Visual Task Transfer](https://mlsanity.com/paper/2408.03326/) (2024)
- [Video-MME: The First-Ever Comprehensive Evaluation Benchmark of Multi-modal LLMs in Video Analysis](https://mlsanity.com/paper/2405.21075/) (2024)
- [MVBench: A Comprehensive Multi-modal Video Understanding Benchmark](https://mlsanity.com/paper/2311.17005/) (2023)
- [MiniCPM-V: A GPT-4V Level MLLM on Your Phone](https://mlsanity.com/paper/2408.01800/) (2024)
- [Dense-Captioning Events in Videos](https://mlsanity.com/paper/1705.00754/) (2017)
- [VideoLLaMA 2: Advancing Spatial-Temporal Modeling and Audio Understanding in Video-LLMs](https://mlsanity.com/paper/2406.07476/) (2024)
- [ActivityNet-QA: A Dataset for Understanding Complex Web Videos via Question Answering](https://mlsanity.com/paper/1906.02467/) (2019)
- [LLaMA-VID: An Image is Worth 2 Tokens in Large Language Models](https://mlsanity.com/paper/2311.17043/) (2023)
- [MovieChat: From Dense Token to Sparse Memory for Long Video Understanding](https://mlsanity.com/paper/2307.16449/) (2023)
- [VideoLLaMA 3: Frontier Multimodal Foundation Models for Image and Video Understanding](https://mlsanity.com/paper/2501.13106/) (2025)
- [Long Context Transfer from Language to Vision](https://mlsanity.com/paper/2406.16852/) (2024)
- [TVQA: Localized, Compositional Video Question Answering](https://mlsanity.com/paper/1809.01696/) (2018)
- [ShareGPT4Video: Improving Video Understanding and Generation with Better Captions](https://mlsanity.com/paper/2406.04325/) (2024)
- [TGIF-QA: Toward Spatio-Temporal Reasoning in Visual Question Answering](https://mlsanity.com/paper/1704.04497/) (2017)
- [Chat-UniVi: Unified Visual Representation Empowers Large Language Models with Image and Video Understanding](https://mlsanity.com/paper/2311.08046/) (2023)
- [LongVU: Spatiotemporal Adaptive Compression for Long Video-Language Understanding](https://mlsanity.com/paper/2410.17434/) (2024)
- [OBELICS: An Open Web-Scale Filtered Dataset of Interleaved Image-Text Documents](https://mlsanity.com/paper/2306.16527/) (2023)
- [LongVILA: Scaling Long-Context Visual Language Models for Long Videos](https://mlsanity.com/paper/2408.10188/) (2024)
- [MovieQA: Understanding Stories in Movies through Question-Answering](https://mlsanity.com/paper/1512.02902/) (2015)
- [mPLUG-Owl3: Towards Long Image-Sequence Understanding in Multi-Modal Large Language Models](https://mlsanity.com/paper/2408.04840/) (2024)
- [VideoChat-Flash: Hierarchical Compression for Long-Context Video Modeling](https://mlsanity.com/paper/2501.00574/) (2024)
- [InternVideo2.5: Empowering Video MLLMs with Long and Rich Context Modeling](https://mlsanity.com/paper/2501.12386/) (2025)
- [Direct Preference Optimization of Video Large Multimodal Models from Language Model Reward](https://mlsanity.com/paper/2404.01258/) (2024)

## Built on by (most cited first, 7 of 7)

- [LOVE-R1: Advancing Long Video Understanding with an Adaptive Zoom-in Mechanism via Multi-Step Reasoning](https://mlsanity.com/paper/2509.24786/) (2025)
- [Kling-Avatar: Grounding Multimodal Instructions for Cascaded Long-Duration Avatar Animation Synthesis](https://mlsanity.com/paper/2509.09595/) (2025)
- [Select Less, Reason More: Prioritizing Evidence Purity for Video Reasoning](https://mlsanity.com/paper/2510.15440/) (2025)
- [Keep What Audio Cannot Say: Context-Preserving Token Pruning for Omni-LLMs](https://mlsanity.com/paper/2605.11605/) (2026)
- [How Well Can Your Video Model Remember? Measuring Memory-Budget Trade-offs in Long Video Understanding](https://mlsanity.com/paper/2606.20726/) (2026)
- [YingVideo-MV: Music-Driven Multi-Stage Video Generation](https://mlsanity.com/paper/2512.02492/) (2025)
- [Why Is Video Still So Expensive? A Survey of Inference-Efficiency Mechanisms in Video and Audiovisual LLMs](https://mlsanity.com/paper/2609.10355/) (2026)

## Abstract (arXiv)

Large Multimodal Models (LMMs) uniformly perceive video frames, creating computational inefficiency for videos with inherently varying temporal information density. This paper present Quicksviewer, an LMM with new perceiving paradigm that partitions a video of nonuniform density into varying cubes using Gumbel Softmax, followed by a unified resampling for each cube to achieve efficient video understanding. This simple and intuitive approach dynamically compress video online based on its temporal density, significantly reducing spatiotemporal redundancy (overall 45× compression rate), while enabling efficient training with large receptive field. We train the model from a language backbone through three progressive stages, each incorporating lengthy videos on average of 420s/1fps thanks to the perceiving efficiency. With only 0.8M total video-text samples for training, our model outperforms the direct baseline employing a fixed partitioning strategy by a maximum of 8.72 in accuracy, demonstrating the effectiveness in performance. On Video-MME, Quicksviewer achieves SOTA under modest sequence lengths using just up to 5% of tokens per frame required by baselines. With this paradigm, scaling up the number of input frames reveals a clear power law of the model capabilities. It is also empirically verified that the segments generated by the cubing network can help for analyzing continuous events in videos.