# Understanding Attention Mechanism in Video Diffusion Models

arXiv: https://arxiv.org/abs/2504.12027 · HTML: https://mlsanity.com/paper/2504.12027/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- Placed in Video generation, one of 14 directions in Diffusion models.
- Builds on 39 of its 46 references that are in the map.
- First posted on arXiv 16 Apr 2025.

## Placed in

- Diffusion models / [Video generation](https://mlsanity.com/diffusion/video-generation/)

## Builds on (most cited first, 30 of 39)

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [Denoising Diffusion Probabilistic Models](https://mlsanity.com/paper/2006.11239/) (2020)
- [High-Resolution Image Synthesis with Latent Diffusion Models](https://mlsanity.com/paper/2112.10752/) (2021)
- [The Unreasonable Effectiveness of Deep Features as a Perceptual Metric](https://mlsanity.com/paper/1801.03924/) (2018)
- [Denoising Diffusion Implicit Models](https://mlsanity.com/paper/2010.02502/) (2020)
- [Scalable Diffusion Models with Transformers](https://mlsanity.com/paper/2212.09748/) (2022)
- [Classifier-Free Diffusion Guidance](https://mlsanity.com/paper/2207.12598/) (2022)
- [LAION-5B: An open large-scale dataset for training next generation image-text models](https://mlsanity.com/paper/2210.08402/) (2022)
- [Prompt-to-Prompt Image Editing with Cross Attention Control](https://mlsanity.com/paper/2208.01626/) (2022)
- [CogVideoX: Text-to-Video Diffusion Models with An Expert Transformer](https://mlsanity.com/paper/2408.06072/) (2024)
- [VBench: Comprehensive Benchmark Suite for Video Generative Models](https://mlsanity.com/paper/2311.17982/) (2023)
- [AnimateDiff: Animate Your Personalized Text-to-Image Diffusion Models without Specific Tuning](https://mlsanity.com/paper/2307.04725/) (2023)
- [CogVideo: Large-scale Pretraining for Text-to-Video Generation via Transformers](https://mlsanity.com/paper/2205.15868/) (2022)
- [Tune-A-Video: One-Shot Tuning of Image Diffusion Models for Text-to-Video Generation](https://mlsanity.com/paper/2212.11565/) (2022)
- [The 2017 DAVIS Challenge on Video Object Segmentation](https://mlsanity.com/paper/1704.00675/) (2017)
- [Plug-and-Play Diffusion Features for Text-Driven Image-to-Image Translation](https://mlsanity.com/paper/2211.12572/) (2022)
- [MasaCtrl: Tuning-Free Mutual Self-Attention Control for Consistent Image Synthesis and Editing](https://mlsanity.com/paper/2304.08465/) (2023)
- [VideoCrafter2: Overcoming Data Limitations for High-Quality Video Diffusion Models](https://mlsanity.com/paper/2401.09047/) (2024)
- [FateZero: Fusing Attentions for Zero-shot Text-based Video Editing](https://mlsanity.com/paper/2303.09535/) (2023)
- [TokenFlow: Consistent Diffusion Features for Consistent Video Editing](https://mlsanity.com/paper/2307.10373/) (2023)
- [LAVIE: High-Quality Video Generation with Cascaded Latent Diffusion Models](https://mlsanity.com/paper/2309.15103/) (2023)
- [ControlVideo: Training-free Controllable Text-to-Video Generation](https://mlsanity.com/paper/2305.13077/) (2023)
- [Pix2Video: Video Editing using Image Diffusion](https://mlsanity.com/paper/2303.12688/) (2023)
- [Video-P2P: Video Editing with Cross-attention Control](https://mlsanity.com/paper/2303.04761/) (2023)
- [Hunyuan-DiT: A Powerful Multi-Resolution Diffusion Transformer with Fine-Grained Chinese Understanding](https://mlsanity.com/paper/2405.08748/) (2024)
- [Rerender A Video: Zero-Shot Text-Guided Video-to-Video Translation](https://mlsanity.com/paper/2306.07954/) (2023)
- [FreeU: Free Lunch in Diffusion U-Net](https://mlsanity.com/paper/2309.11497/) (2023)
- [Interpreting CLIP's Image Representation via Text-Based Decomposition](https://mlsanity.com/paper/2310.05916/) (2023)
- [FLATTEN: optical FLow-guided ATTENtion for consistent text-to-video editing](https://mlsanity.com/paper/2310.05922/) (2023)
- [AnyV2V: A Tuning-Free Framework For Any Video-to-Video Editing Tasks](https://mlsanity.com/paper/2403.14468/) (2024)

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

- [Efficient Video Diffusion Models: Advancements and Challenges](https://mlsanity.com/paper/2604.15911/) (2026)
- [CLEAR: Context-Aware Learning with End-to-End Mask-Free Inference for Adaptive Video Subtitle Removal](https://mlsanity.com/paper/2603.21901/) (2026)
- [Why Do Video Diffusion Models Violate Physics? Unveiling the Flaws in Attention Mechanisms](https://mlsanity.com/paper/2609.23658/) (2026)
- [Zero-to-Hero: Zero-Shot Initialization Empowering Reference-Based Video Appearance Editing](https://mlsanity.com/paper/2505.23134/) (2025)
- [QSilk: Micrograin Stabilization and Adaptive Quantile Clipping for Detail-Friendly Latent Diffusion](https://mlsanity.com/paper/2510.15761/) (2025)
- [Controlling Motion Transfer in Diffusion Transformers via Attention Heads](https://mlsanity.com/paper/2607.11081/) (2026)
- [Temporal-Attention Head Specialization During Video Diffusion Training](https://mlsanity.com/paper/2609.31654/) (2026)

## Abstract (arXiv)

Text-to-video (T2V) synthesis models, such as OpenAI's Sora, have garnered significant attention due to their ability to generate high-quality videos from a text prompt. In diffusion-based T2V models, the attention mechanism is a critical component. However, it remains unclear what intermediate features are learned and how attention blocks in T2V models affect various aspects of video synthesis, such as image quality and temporal consistency. In this paper, we conduct an in-depth perturbation analysis of the spatial and temporal attention blocks of T2V models using an information-theoretic approach. Our results indicate that temporal and spatial attention maps affect not only the timing and layout of the videos but also the complexity of spatiotemporal elements and the aesthetic quality of the synthesized videos. Notably, high-entropy attention maps are often key elements linked to superior video quality, whereas low-entropy attention maps are associated with the video's intra-frame structure. Based on our findings, we propose two novel methods to enhance video quality and enable text-guided video editing. These methods rely entirely on lightweight manipulation of the attention matrices in T2V models. The efficacy and effectiveness of our methods are further validated through experimental evaluation across multiple datasets.