# LGTM: Training-Free Light-Guided Text-to-Image Diffusion Model via Initial Noise Manipulation

arXiv: https://arxiv.org/abs/2603.24086 · HTML: https://mlsanity.com/paper/2603.24086/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Placed in Image generation, text-to-image and editing, one of 14 directions in Diffusion models.
- Builds on 38 of its 46 references that are in the map.
- First posted on arXiv 25 Mar 2026.

## Placed in

- Diffusion models / [Image generation, text-to-image and editing](https://mlsanity.com/diffusion/image-generation-text-to-image-and-editing/)

## Builds on (most cited first, 30 of 38)

- [Denoising Diffusion Probabilistic Models](https://mlsanity.com/paper/2006.11239/) (2020)
- [High-Resolution Image Synthesis with Latent Diffusion Models](https://mlsanity.com/paper/2112.10752/) (2021)
- [Denoising Diffusion Implicit Models](https://mlsanity.com/paper/2010.02502/) (2020)
- [Photorealistic Text-to-Image Diffusion Models with Deep Language Understanding](https://mlsanity.com/paper/2205.11487/) (2022)
- [Adding Conditional Control to Text-to-Image Diffusion Models](https://mlsanity.com/paper/2302.05543/) (2023)
- [SDXL: Improving Latent Diffusion Models for High-Resolution Image Synthesis](https://mlsanity.com/paper/2307.01952/) (2023)
- [BLIP: Bootstrapping Language-Image Pre-training for Unified Vision-Language Understanding and Generation](https://mlsanity.com/paper/2201.12086/) (2022)
- [CLIPScore: A Reference-free Evaluation Metric for Image Captioning](https://mlsanity.com/paper/2104.08718/) (2021)
- [eDiff-I: Text-to-Image Diffusion Models with an Ensemble of Expert Denoisers](https://mlsanity.com/paper/2211.01324/) (2022)
- [Attend-and-Excite: Attention-Based Semantic Guidance for Text-to-Image Diffusion Models](https://mlsanity.com/paper/2301.13826/) (2023)
- [MultiDiffusion: Fusing Diffusion Paths for Controlled Image Generation](https://mlsanity.com/paper/2302.08113/) (2023)
- [Make-A-Scene: Scene-Based Text-to-Image Generation with Human Priors](https://mlsanity.com/paper/2203.13131/) (2022)
- [NIMA: Neural Image Assessment](https://mlsanity.com/paper/1709.05424/) (2017)
- [LayoutDiffusion: Controllable Diffusion Model for Layout-to-image Generation](https://mlsanity.com/paper/2303.17189/) (2023)
- [Design Guidelines for Prompt Engineering Text-to-Image Generative Models](https://mlsanity.com/paper/2109.06977/) (2021)
- [ReCo: Region-Controlled Text-to-Image Generation](https://mlsanity.com/paper/2211.15518/) (2022)
- [ReNO: Enhancing One-step Text-to-Image Models through Reward-based Noise Optimization](https://mlsanity.com/paper/2406.04312/) (2024)
- [DiLightNet: Fine-grained Lighting Control for Diffusion-based Image Generation](https://mlsanity.com/paper/2402.11929/) (2024)
- [FreeControl: Training-Free Spatial Control of Any Text-to-Image Diffusion Model with Any Condition](https://mlsanity.com/paper/2312.07536/) (2023)
- [Not All Noises Are Created Equally:Diffusion Noise Selection and Optimization](https://mlsanity.com/paper/2407.14041/) (2024)
- [Relightful Harmonization: Lighting-aware Portrait Background Replacement](https://mlsanity.com/paper/2312.06886/) (2023)
- [LightIt: Illumination Modeling and Control for Diffusion Models](https://mlsanity.com/paper/2403.10615/) (2024)
- [FlashTex: Fast Relightable Mesh Texturing with LightControlNet](https://mlsanity.com/paper/2402.13251/) (2024)
- [DesignDiffusion: High-Quality Text-to-Design Image Generation with Diffusion Models](https://mlsanity.com/paper/2503.01645/) (2025)
- [Noise Hypernetworks: Amortizing Test-Time Compute in Diffusion Models](https://mlsanity.com/paper/2508.09968/) (2025)
- [All Seeds Are Not Equal: Enhancing Compositional Text-to-Image Generation with Reliable Random Seeds](https://mlsanity.com/paper/2411.18810/) (2024)
- [It's All About Your Sketch: Democratising Sketch Control in Diffusion Models](https://mlsanity.com/paper/2403.07234/) (2024)
- [The Lottery Ticket Hypothesis in Denoising: Towards Semantic-Driven Initialization](https://mlsanity.com/paper/2312.08872/) (2023)
- [The Crystal Ball Hypothesis in diffusion models: Anticipating object positions from initial noise](https://mlsanity.com/paper/2406.01970/) (2024)
- [Training-Free Location-Aware Text-to-Image Synthesis](https://mlsanity.com/paper/2304.13427/) (2023)

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

- [LimeCross: Context-Conditioned Layered Image Editing with Structural Consistency](https://mlsanity.com/paper/2605.10319/) (2026)

## Abstract (arXiv)

Diffusion models have demonstrated high-quality performance in conditional text-to-image generation, particularly with structural cues such as edges, layouts, and depth. However, lighting conditions have received limited attention and remain difficult to control within the generative process. Existing methods handle lighting through a two-stage pipeline that relights images after generation, which is inefficient. Moreover, they rely on fine-tuning with large datasets and heavy computation, limiting their adaptability to new models and tasks. To address this, we propose a novel Training-Free Light-Guided Text-to-Image Diffusion Model via Initial Noise Manipulation (LGTM), which manipulates the initial latent noise of the diffusion process to guide image generation with text prompts and user-specified light directions. Through a channel-wise analysis of the latent space, we find that selectively manipulating latent channels enables fine-grained lighting control without fine-tuning or modifying the pre-trained model. Extensive experiments show that our method surpasses prompt-based baselines in lighting consistency, while preserving image quality and text alignment. This approach introduces new possibilities for dynamic, user-guided light control. Furthermore, it integrates seamlessly with models like ControlNet, demonstrating adaptability across diverse scenarios.