# All modalities are equal, but video is more equal: Closing the Cross-Attention Gap in Joint Video Generation

arXiv: https://arxiv.org/abs/2609.27901 · HTML: https://mlsanity.com/paper/2609.27901/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 56 of its 62 references that are in the map.
- First posted on arXiv 23 Sep 2026.

## Builds on (most cited first, 30 of 56)

- [Adding Conditional Control to Text-to-Image Diffusion Models](https://mlsanity.com/paper/2302.05543/) (2023)
- [Prompt-to-Prompt Image Editing with Cross Attention Control](https://mlsanity.com/paper/2208.01626/) (2022)
- [VBench: Comprehensive Benchmark Suite for Video Generative Models](https://mlsanity.com/paper/2311.17982/) (2023)
- [Zero-1-to-3: Zero-shot One Image to 3D Object](https://mlsanity.com/paper/2303.11328/) (2023)
- [IP-Adapter: Text Compatible Image Prompt Adapter for Text-to-Image Diffusion Models](https://mlsanity.com/paper/2308.06721/) (2023)
- [Attend-and-Excite: Attention-Based Semantic Guidance for Text-to-Image Diffusion Models](https://mlsanity.com/paper/2301.13826/) (2023)
- [Keep it SMPL: Automatic Estimation of 3D Human Pose and Shape from a Single Image](https://mlsanity.com/paper/1607.08128/) (2016)
- [VGGSound: A Large-scale Audio-Visual Dataset](https://mlsanity.com/paper/2004.14368/) (2020)
- [MediaPipe: A Framework for Building Perception Pipelines](https://mlsanity.com/paper/1906.08172/) (2019)
- [What Makes Training Multi-Modal Classification Networks Hard?](https://mlsanity.com/paper/1905.12681/) (2019)
- [VBench-2.0: Advancing Video Generation Benchmark Suite for Intrinsic Faithfulness](https://mlsanity.com/paper/2503.21755/) (2025)
- [Balanced Multimodal Learning via On-the-fly Gradient Modulation](https://mlsanity.com/paper/2203.15332/) (2022)
- [LTX-2: Efficient Joint Audio-Visual Foundation Model](https://mlsanity.com/paper/2601.03233/) (2026)
- [MM-Diffusion: Learning Multi-Modal Diffusion Models for Joint Audio and Video Generation](https://mlsanity.com/paper/2212.09478/) (2022)
- [Taming Visually Guided Sound Generation](https://mlsanity.com/paper/2110.08791/) (2021)
- [Diff-Foley: Synchronized Video-to-Audio Synthesis with Latent Diffusion Models](https://mlsanity.com/paper/2306.17203/) (2023)
- [MMAudio: Taming Multimodal Joint Training for High-Quality Video-to-Audio Synthesis](https://mlsanity.com/paper/2412.15322/) (2024)
- [VidProM: A Million-scale Real Prompt-Gallery Dataset for Text-to-Video Diffusion Models](https://mlsanity.com/paper/2403.06098/) (2024)
- [PMR: Prototypical Modal Rebalance for Multimodal Learning](https://mlsanity.com/paper/2211.07089/) (2022)
- [VideoJAM: Joint Appearance-Motion Representations for Enhanced Motion Generation in Video Models](https://mlsanity.com/paper/2502.02492/) (2025)
- [FoleyCrafter: Bring Silent Videos to Life with Lifelike and Synchronized Sounds](https://mlsanity.com/paper/2407.01494/) (2024)
- [Synchformer: Efficient Synchronization from Sparse Cues](https://mlsanity.com/paper/2401.16423/) (2024)
- [I Hear Your True Colors: Image Guided Audio Generation](https://mlsanity.com/paper/2211.03089/) (2022)
- [Ovi: Twin Backbone Cross-Modal Fusion for Audio-Video Generation](https://mlsanity.com/paper/2510.01284/) (2025)
- [Diverse and Aligned Audio-to-Video Generation via Text-to-Video Model Adaptation](https://mlsanity.com/paper/2309.16429/) (2023)
- [CameraHMR: Aligning People with Perspective](https://mlsanity.com/paper/2411.08128/) (2024)
- [UniVerse-1: Unified Audio-Video Generation via Stitching of Experts](https://mlsanity.com/paper/2509.06155/) (2025)
- [Removing Bias in Multi-modal Classifiers: Regularization by Maximizing Functional Entropies](https://mlsanity.com/paper/2010.10802/) (2020)
- [MOVA: Towards Scalable and Synchronized Video-Audio Generation](https://mlsanity.com/paper/2602.08794/) (2026)
- [Optimizing Relevance Maps of Vision Transformers Improves Robustness](https://mlsanity.com/paper/2206.01161/) (2022)

## Abstract (arXiv)

Video is a rich representation of a physical event, capturing appearance, geometry, motion, and temporal evolution. Other modalities, such as 3D body motion or audio, encode narrower aspects of the same event. We find that joint multimodal diffusion transformers exhibit a corresponding asymmetry in cross-modal correspondence: companion modalities develop strong correspondences to video, but the reciprocal correspondences through which they constrain video remain substantially weaker. We express both directions as comparable correspondence distributions over video tokens and define their disagreement as the reciprocal correspondence gap. We introduce RecCAR, standing for Reciprocal Cross-modal Attention Regularization, a KL regularizer that uses the well-established video-to-modality correspondence as a fixed reference and aligns the weaker modality-to-video correspondence toward it. Across joint video-motion and video-audio generation, RecCAR improves the Human Anatomy score from 0.69 to 0.75 and reduces audio-video desynchronization from 0.804 to 0.752, while improving overall generation