mlsanityData to 14 Oct 2026

First Frame Is the Place to Go for Video Content Customization

arXiv 2511.15700 ↗First posted 19 Nov 2025cs.CVcited by 12

Not on the map. Why?

Abstract (arXiv)

What role does the first frame play in video generation models? Traditionally, it's viewed as the spatial-temporal starting point of a video, merely a seed for subsequent animation. In this work, we reveal a fundamentally different perspective: video models implicitly treat the first frame as a conceptual memory buffer that stores visual entities for later reuse during generation. Leveraging this insight, we show that it's possible to achieve robust and generalized video content customization in diverse scenarios, using only 20-50 training examples without architectural changes or large-scale finetuning. This unveils a powerful, overlooked capability of video generation models for reference-based video customization.

Key facts. Cited by 12 papers in the map (data through 14 Oct 2026). Builds on 43 of its 49 references that are in the map. First posted on arXiv 19 Nov 2025.

Builds on

43 of 49 references.

References · 43

  1. 2021High-Resolution Image Synthesis with Latent Diffusion Modelscited by 18,863
  2. 2022Scalable Diffusion Models with Transformerscited by 6,054
  3. 2024GPT-4o System Cardcited by 4,588
  4. 2025Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilitiescited by 3,796
  5. 2022DreamBooth: Fine Tuning Text-to-Image Diffusion Models for Subject-Driven Generationcited by 3,300
  6. 2024SAM 2: Segment Anything in Images and Videoscited by 2,863
  7. 2025Wan: Open and Advanced Large-Scale Video Generative Modelscited by 2,708
  8. 2023Stable Video Diffusion: Scaling Latent Video Diffusion Models to Large Datasetscited by 2,581
  9. 2024CogVideoX: Text-to-Video Diffusion Models with An Expert Transformercited by 2,261
  10. 2023MMBench: Is Your Multi-modal Model an All-around Player?cited by 2,141
  11. 2017Decoupled Weight Decay Regularizationcited by 1,831
  12. 2025Qwen-Image Technical Reportcited by 1,044
31 more
  1. 2024Emu3: Next-Token Prediction is All You Needcited by 781
  2. 2025Qwen2.5-Omni Technical Reportcited by 736
  3. 2024SpatialVLM: Endowing Vision-Language Models with Spatial Reasoning Capabilitiescited by 706
  4. 2023Baichuan 2: Open Large-scale Language Modelscited by 670
  5. 2024Movie Gen: A Cast of Media Foundation Modelscited by 644
  6. 2025Video-R1: Reinforcing Video Reasoning in MLLMscited by 484
  7. 2024Transfusion: Predict the Next Token and Diffuse Images with One Multi-Modal Modelcited by 483
  8. 2025VACE: All-in-One Video Creation and Editingcited by 479
  9. 2024Sora: A Review on Background, Technology, Limitations, and Opportunities of Large Vision Modelscited by 477
  10. 2023Mitigating Hallucination in Large Multi-Modal Models via Robust Instruction Tuningcited by 445
  11. 2023Vid2Seq: Large-Scale Pretraining of a Visual Language Model for Dense Video Captioningcited by 263
  12. 2025Video models are zero-shot learners and reasonerscited by 220
  13. 2024PhysDreamer: Physics-Based Interaction with 3D Objects via Video Generationcited by 191
  14. 2023DreamVideo: Composing Your Dream Videos with Customized Subject and Motioncited by 188
  15. 2024In-Context LoRA for Diffusion Transformerscited by 158
  16. 2024Identity-Preserving Text-to-Video Generation by Frequency Decompositioncited by 158
  17. 2024MS-Diffusion: Multi-subject Zero-shot Image Personalization with Layout Guidancecited by 144
  18. 2024Eagle: Exploring The Design Space for Multimodal LLMs with Mixture of Encoderscited by 139
  19. 2023VideoBooth: Diffusion-based Video Generation with Image Promptscited by 126
  20. 2025Go-with-the-Flow: Motion-Controllable Video Diffusion Models Using Real-Time Warped Noisecited by 117
  21. 2025Self-Rewarding Vision-Language Model via Reasoning Decompositioncited by 97
  22. 2025SkyReels-A2: Compose Anything in Video Diffusion Transformerscited by 73
  23. 2025Multi-subject Open-set Personalization in Video Generationcited by 66
  24. 2023Revisiting Few-Shot Object Detection with Vision-Language Modelscited by 32
  25. 2025VideoHallu: Evaluating and Mitigating Multi-modal Hallucinations on Synthetic Video Understandingcited by 32
  26. 2025HOIGen-1M: A Large-scale Dataset for Human-Object Interaction Video Generationcited by 26
  27. 2024Repurposing Pre-trained Video Diffusion Models for Event-based Video Interpolationcited by 23
  28. 2025DreamActor-H1: High-Fidelity Human-Product Demonstration Video Generation via Motion-designed Diffusion Transformerscited by 22
  29. 2024Distilling Vision-Language Models on Millions of Videoscited by 20
  30. 2025VideoMage: Multi-Subject and Motion Customization of Text-to-Video Diffusion Modelscited by 17
  31. 2025Character Mixing for Video Generationcited by 2

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 1; 2026: 11.

12 papers.

  1. 2026VideoGPA: Distilling Geometry Priors for 3D-Consistent Video Generationcited by 23
  2. 2026Echo-Memory: A Controlled Study of Memory in Action World Modelscited by 6
  3. 2025From Inpainting to Layer Decomposition: Repurposing Generative Inpainting Models for Image Layer Decompositioncited by 4
  4. 2026Omni-Customizer: End-to-End MultiModal Customization for Joint Audio-Video Generationcited by 2
  5. 2026OmniTryOn: Video Try-On Anything at Once!cited by 1
  6. 2026DomainShuttle: Freeform Open Domain Subject-driven Text-to-video Generationcited by 1
  7. 2026Apple-π: Benchmarking Thinking with Video Towards Law-Grounded Physical Intelligencecited by 1
  8. 2026HOMIE: Human-object Centric Video Personalization via Multimodal Intelligent Enhancementcited by 1
  9. 2026Spatial-Temporal Decoupled Reference Conditioning for Identity-Preserving Text-to-Video Generationcited by 0
  10. 2026RhymeFlow: Training-Free Acceleration for Video Generation with Asynchronous Denoising Flow Schedulingcited by 0
  11. 2026Golden Anchor for Video Generationcited by 0
  12. 2026CineMem: Attention-Aligned Fine-grained Memory Retrieval for Multi-Shot Long Video Generationcited by 0