# LumiTokens: 3D Relighting via Token-Space Lighting Transformation

arXiv: https://arxiv.org/abs/2608.18215 · HTML: https://mlsanity.com/paper/2608.18215/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 36 of its 41 references that are in the map.
- First posted on arXiv 18 Aug 2026.

## Builds on (most cited first, 30 of 36)

- [The Unreasonable Effectiveness of Deep Features as a Perceptual Metric](https://mlsanity.com/paper/1801.03924/) (2018)
- [3D Gaussian Splatting for Real-Time Radiance Field Rendering](https://mlsanity.com/paper/2308.04079/) (2023)
- [Vision Transformers for Dense Prediction](https://mlsanity.com/paper/2103.13413/) (2021)
- [Objaverse: A Universe of Annotated 3D Objects](https://mlsanity.com/paper/2212.08051/) (2022)
- [Efficient Geometry-aware 3D Generative Adversarial Networks](https://mlsanity.com/paper/2112.07945/) (2021)
- [LRM: Large Reconstruction Model for Single Image to 3D](https://mlsanity.com/paper/2311.04400/) (2023)
- [NeRD: Neural Reflectance Decomposition from Image Collections](https://mlsanity.com/paper/2012.03918/) (2020)
- [GS-LRM: Large Reconstruction Model for 3D Gaussian Splatting](https://mlsanity.com/paper/2404.19702/) (2024)
- [PhySG: Inverse Rendering with Spherical Gaussians for Physics-based Material Editing and Relighting](https://mlsanity.com/paper/2104.00674/) (2021)
- [Scene Representation Transformer: Geometry-Free Novel View Synthesis Through Set-Latent Scene Representations](https://mlsanity.com/paper/2111.13152/) (2021)
- [GS-IR: 3D Gaussian Splatting for Inverse Rendering](https://mlsanity.com/paper/2311.16473/) (2023)
- [LVSM: A Large View Synthesis Model with Minimal 3D Inductive Bias](https://mlsanity.com/paper/2410.17242/) (2024)
- [TensoIR: Tensorial Inverse Rendering](https://mlsanity.com/paper/2304.12461/) (2023)
- [RGBleftrightarrowX: Image decomposition and synthesis using material- and lighting-aware diffusion models](https://mlsanity.com/paper/2405.00666/) (2024)
- [Shape, Light, and Material Decomposition from Images using Monte Carlo Rendering and Denoising](https://mlsanity.com/paper/2206.03380/) (2022)
- [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)
- [RayZer: A Self-supervised Large View Synthesis Model](https://mlsanity.com/paper/2505.00702/) (2025)
- [Intrinsic Image Diffusion for Indoor Single-view Material Estimation](https://mlsanity.com/paper/2312.12274/) (2023)
- [DiffusionRenderer: Neural Inverse and Forward Rendering with Video Diffusion Models](https://mlsanity.com/paper/2501.18590/) (2025)
- [I2-SDF: Intrinsic Indoor Scene Reconstruction and Editing via Raytracing in Neural SDFs](https://mlsanity.com/paper/2303.07634/) (2023)
- [UniRelight: Learning Joint Decomposition and Synthesis for Video Relighting](https://mlsanity.com/paper/2506.15673/) (2025)
- [A Diffusion Approach to Radiance Field Relighting using Multi-Illumination Synthesis](https://mlsanity.com/paper/2409.08947/) (2024)
- [IllumiNeRF: 3D Relighting Without Inverse Rendering](https://mlsanity.com/paper/2406.06527/) (2024)
- [RelitLRM: Generative Relightable Radiance for Large Reconstruction Models](https://mlsanity.com/paper/2410.06231/) (2024)
- [E-RayZer: Self-supervised 3D Reconstruction as Spatial Visual Pre-training](https://mlsanity.com/paper/2512.10950/) (2025)
- [GenLit: Reformulating Single-Image Relighting as Video Generation](https://mlsanity.com/paper/2412.11224/) (2024)
- [LightSwitch: Multi-view Relighting with Material-guided Diffusion](https://mlsanity.com/paper/2508.06494/) (2025)
- [SceneTok: A Compressed, Diffusable Token Space for 3D Scenes](https://mlsanity.com/paper/2602.18882/) (2026)
- [LuxRemix: Lighting Decomposition and Remixing for Indoor Scenes](https://mlsanity.com/paper/2601.15283/) (2026)

## Abstract (arXiv)

Existing 3D relighting methods operate through either explicit material decomposition, diffusion-based view-space generation, or a combination of both, requiring full recomputation for each new lighting condition. We observe that recent latent scene representations, which encode multi-view images into a set of compact tokens with no fixed physical semantics, open up a novel design space for relighting. We present LumiTokens, a framework that formulates 3D relighting as a direct transformation on latent scene tokens, without explicit 3D representations, rendering equations, or physics-based decomposition. Our model introduces a Scene Token Editor that processes scene tokens jointly with light-ray tokens through self-attention, producing updated tokens that can be decoded into multi-view-consistent relit images. To support diverse lighting types through a unified interface, all lighting signals, including environment maps, point lights, and area lights, are parameterized as Plucker ray tokens, enabling native 3D user interaction with a representation that carries no explicit spatial structure. Crucially, this design supports progressive relighting: because the editor's output remains in the same latent space as its input, a user can incrementally build up illumination one light source at a time, with each edit composing in token space. Experiments demonstrate that LumiTokens achieves comparable or superior relighting quality to other methods and supports progressive, composable lighting edits. Project page: https://neu-vi.github.io/LumiTokens/