mlsanityData to 14 Oct 2026

Motion Attribution for Video Generation

arXiv 2601.08828 ↗First posted 13 Jan 2026cs.CV cs.AI cs.LGcited by 5

Not on the map. Why?

Abstract (arXiv)

Despite the rapid progress of video generation models, the role of data in influencing motion is poorly understood. We present Motive (MOTIon attribution for Video gEneration), a motion-centric, gradient-based data attribution framework that scales to modern, large, high-quality video datasets and models. We use this to study which fine-tuning clips improve or degrade temporal dynamics. Motive isolates temporal dynamics from static appearance via motion-weighted loss masks, yielding efficient and scalable motion-specific influence computation. On text-to-video models, Motive identifies clips that strongly affect motion and guides data curation that improves temporal consistency and physical plausibility. With Motive-selected high-influence data, our method improves both motion smoothness and dynamic degree on VBench, achieving a 74.1% human preference win rate compared with the pretrained base model. To our knowledge, this is the first framework to attribute motion rather than visual appearance in video generative models and to use it to curate fine-tuning data.

Key facts. Cited by 5 papers in the map (data through 14 Oct 2026). Builds on 70 of its 84 references that are in the map. First posted on arXiv 13 Jan 2026.

Builds on

70 of 84 references.

References · 70

  1. 2020Denoising Diffusion Probabilistic Modelscited by 21,076
  2. 2021High-Resolution Image Synthesis with Latent Diffusion Modelscited by 18,863
  3. 2019Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformercited by 14,599
  4. 2020Scaling Laws for Neural Language Modelscited by 6,586
  5. 2022Photorealistic Text-to-Image Diffusion Models with Deep Language Understandingcited by 6,355
  6. 2022Scalable Diffusion Models with Transformerscited by 6,054
  7. 2022Flow Matching for Generative Modelingcited by 5,702
  8. 2021Improved Denoising Diffusion Probabilistic Modelscited by 3,761
  9. 2025Wan: Open and Advanced Large-Scale Video Generative Modelscited by 2,708
  10. 2023Stable Video Diffusion: Scaling Latent Video Diffusion Models to Large Datasetscited by 2,581
  11. 2020RAFT: Recurrent All-Pairs Field Transforms for Optical Flowcited by 2,415
  12. 2022Video Diffusion Modelscited by 2,305
38 more
  1. 2024CogVideoX: Text-to-Video Diffusion Models with An Expert Transformercited by 2,261
  2. 2023VBench: Comprehensive Benchmark Suite for Video Generative Modelscited by 1,822
  3. 2017Understanding Black-box Predictions via Influence Functionscited by 1,741
  4. 2023AnimateDiff: Animate Your Personalized Text-to-Image Diffusion Models without Specific Tuningcited by 1,433
  5. 2022Tune-A-Video: One-Shot Tuning of Image Diffusion Models for Text-to-Video Generationcited by 961
  6. 2023Stochastic Interpolants: A Unifying Framework for Flows and Diffusionscited by 939
  7. 2025Cosmos World Foundation Model Platform for Physical AIcited by 864
  8. 2025V-JEPA 2: Self-Supervised Video Models Enable Understanding, Prediction and Planningcited by 809
  9. 2024LTX-Video: Realtime Video Latent Diffusioncited by 581
  10. 2020Estimating Training Data Influence by Tracing Gradient Descentcited by 580
  11. 2017MoCoGAN: Decomposing Motion and Content for Video Generationcited by 524
  12. 2020What Neural Networks Memorize and Why: Discovering the Long Tail via Influence Estimationcited by 497
  13. 2023TokenFlow: Consistent Diffusion Features for Consistent Video Editingcited by 442
  14. 2019Optimizing Millions of Hyperparameters by Implicit Differentiationcited by 304
  15. 2023TRAK: Attributing Model Behavior at Scalecited by 283
  16. 2025Video models are zero-shot learners and reasonerscited by 220
  17. 2024How Far is Video Generation from World Model: A Physical Law Perspectivecited by 212
  18. 2024Stable Audio Opencited by 197
  19. 2022Training Data Influence Analysis and Estimation: A Surveycited by 135
  20. 2023Upscale-A-Video: Temporal-Consistent Diffusion Model for Real-World Video Super-Resolutioncited by 132
  21. 2023DataInf: Efficiently Estimating Data Influence in LoRA-tuned LLMs and Diffusion Modelscited by 123
  22. 2018The Mechanics of n-Player Differentiable Gamescited by 118
  23. 2023VideoLCM: Video Latent Consistency Modelcited by 92
  24. 2024Data Shapley in One Training Runcited by 81
  25. 2014Fastfood: Approximate Kernel Expansions in Loglinear Timecited by 74
  26. 2018Stochastic Hyperparameter Optimization through Hypernetworkscited by 73
  27. 2024VidGen-1M: A Large-Scale Dataset for Text-to-video Generationcited by 62
  28. 2023Controllable Text-to-Image Generation with GPT-4cited by 59
  29. 2024COVE: Unleashing the Diffusion Feature Correspondence for Consistent Video Editingcited by 59
  30. 2025AllTracker: Efficient Dense Point Tracking at High Resolutioncited by 58
  31. 2023Control-A-Video: Controllable Text-to-Video Diffusion Models with Motion Prior and Reward Feedback Learningcited by 48
  32. 2023Intriguing Properties of Data Attribution on Diffusion Modelscited by 47
  33. 2023Evaluating Data Attribution for Text-to-Image Modelscited by 45
  34. 2023The Journey, Not the Destination: How Data Guides Diffusion Modelscited by 43
  35. 2024EmerDiff: Emerging Pixel-level Semantic Knowledge in Diffusion Modelscited by 43
  36. 20254DNeX: Feed-Forward 4D Generative Modeling Made Easycited by 34
  37. 2024Influence Functions for Scalable Data Attribution in Diffusion Modelscited by 33
  38. 2025Astra: General Interactive World Model with Autoregressive Denoisingcited by 33

50 most cited of 70 All 70 (JSON)

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 5.

5 papers.

  1. 2026Addressable Memory for Video World Modelscited by 15
  2. 2026LaMo: Self-Supervised Latent Motion Priors for Physical Realism in Video Generationcited by 4
  3. 2026World-Ego Modeling for Long-Horizon Evolution in Hybrid Embodied Taskscited by 2
  4. 2026Variance Reduction for Expectations with Diffusion Teacherscited by 2
  5. 2026VidaForge: Open Research Infrastructure for Video Pretraining Data Recipescited by 0