# HorizonRelight: Relighting Long-horizon Videos Consistently via Diffusion Transformers

arXiv: https://arxiv.org/abs/2606.29095 · HTML: https://mlsanity.com/paper/2606.29095/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Placed in 3D, point cloud and scene generation, one of 14 directions in Diffusion models.
- Builds on 25 of its 31 references that are in the map.
- First posted on arXiv 27 Jun 2026.

## Placed in

- Diffusion models / [3D, point cloud and scene generation](https://mlsanity.com/diffusion/3d-point-cloud-and-scene-generation/)

## Builds on (most cited first, 25 of 25)

- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [GPT-4 Technical Report](https://mlsanity.com/paper/2303.08774/) (2023)
- [Scalable Diffusion Models with Transformers](https://mlsanity.com/paper/2212.09748/) (2022)
- [Prefix-Tuning: Optimizing Continuous Prompts for Generation](https://mlsanity.com/paper/2101.00190/) (2021)
- [The Power of Scale for Parameter-Efficient Prompt Tuning](https://mlsanity.com/paper/2104.08691/) (2021)
- [Elucidating the Design Space of Diffusion-Based Generative Models](https://mlsanity.com/paper/2206.00364/) (2022)
- [Stable Video Diffusion: Scaling Latent Video Diffusion Models to Large Datasets](https://mlsanity.com/paper/2311.15127/) (2023)
- [Cosmos World Foundation Model Platform for Physical AI](https://mlsanity.com/paper/2501.03575/) (2025)
- [Latte: Latent Diffusion Transformer for Video Generation](https://mlsanity.com/paper/2401.03048/) (2024)
- [RGBleftrightarrowX: Image decomposition and synthesis using material- and lighting-aware diffusion models](https://mlsanity.com/paper/2405.00666/) (2024)
- [VideoJAM: Joint Appearance-Motion Representations for Enhanced Motion Generation in Video Models](https://mlsanity.com/paper/2502.02492/) (2025)
- [Cosmos-Transfer1: Conditional World Generation with Adaptive Multimodal Control](https://mlsanity.com/paper/2503.14492/) (2025)
- [VSA: Faster Video Diffusion with Trainable Sparse Attention](https://mlsanity.com/paper/2505.13389/) (2025)
- [DiLightNet: Fine-grained Lighting Control for Diffusion-based Image Generation](https://mlsanity.com/paper/2402.11929/) (2024)
- [Neural Gaffer: Relighting Any Object via Diffusion](https://mlsanity.com/paper/2406.07520/) (2024)
- [Long-Context State-Space Video World Models](https://mlsanity.com/paper/2505.20171/) (2025)
- [LightIt: Illumination Modeling and Control for Diffusion Models](https://mlsanity.com/paper/2403.10615/) (2024)
- [DiffusionRenderer: Neural Inverse and Forward Rendering with Video Diffusion Models](https://mlsanity.com/paper/2501.18590/) (2025)
- [Deep Shading: Convolutional Neural Networks for Screen-Space Shading](https://mlsanity.com/paper/1603.06078/) (2016)
- [Matrix3D: Large Photogrammetry Model All-in-One](https://mlsanity.com/paper/2502.07685/) (2025)
- [UniRelight: Learning Joint Decomposition and Synthesis for Video Relighting](https://mlsanity.com/paper/2506.15673/) (2025)
- [RelitLRM: Generative Relightable Radiance for Large Reconstruction Models](https://mlsanity.com/paper/2410.06231/) (2024)
- [GenLit: Reformulating Single-Image Relighting as Video Generation](https://mlsanity.com/paper/2412.11224/) (2024)
- [DCP: Addressing Input Dynamism In Long-Context Training via Dynamic Context Parallelism](https://mlsanity.com/paper/2510.10620/) (2025)

## Built on by (most cited first, 1 of 1)

- [Generative World Renderer at the Speed of Play](https://mlsanity.com/paper/2607.18703/) (2026)

## Abstract (arXiv)

Diffusion-based video relighting enables controllable relighting from a single input video, but modern video diffusion backbones are trained on short clips and applied to long-horizon videos through chunked sliding-window inference, often causing temporal discontinuities at chunk boundaries. We address this by reframing long-horizon relighting as temporally conditioned latent domain translation. Our framework enforces cross-chunk continuity by propagating target-domain latents across boundaries and makes this behavior learnable using masked target-domain self-conditioning, training the model to continue from temporally masked propagated context. We further introduce warm-start prompting with a relit prompt anchor from a controllable generative model, which establishes the initial target-domain state and creates a general interface for prompt-based relighting. Experiments on in-the-wild long-horizon videos show markedly improved temporal consistency, with chunk-boundary artifacts largely reduced and unwanted appearance changes across chunks greatly suppressed.