# Debiasing Diffusion Priors via 3D Attention for Consistent Gaussian Splatting arXiv: https://arxiv.org/abs/2512.07345 ยท HTML: https://mlsanity.com/paper/2512.07345/ ## 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 27 of its 41 references that are in the map. - First posted on arXiv 8 Dec 2025. ## 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, 27 of 27) - [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021) - [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020) - [U-Net: Convolutional Networks for Biomedical Image Segmentation](https://mlsanity.com/paper/1505.04597/) (2015) - [3D Gaussian Splatting for Real-Time Radiance Field Rendering](https://mlsanity.com/paper/2308.04079/) (2023) - [DreamFusion: Text-to-3D using 2D Diffusion](https://mlsanity.com/paper/2209.14988/) (2022) - [Zero-1-to-3: Zero-shot One Image to 3D Object](https://mlsanity.com/paper/2303.11328/) (2023) - [MVDream: Multi-view Diffusion for 3D Generation](https://mlsanity.com/paper/2308.16512/) (2023) - [DreamGaussian: Generative Gaussian Splatting for Efficient 3D Content Creation](https://mlsanity.com/paper/2309.16653/) (2023) - [Score Jacobian Chaining: Lifting Pretrained 2D Diffusion Models for 3D Generation](https://mlsanity.com/paper/2212.00774/) (2022) - [Zero-Shot Text-Guided Object Generation with Dream Fields](https://mlsanity.com/paper/2112.01455/) (2021) - [Latent-NeRF for Shape-Guided Generation of 3D Shapes and Textures](https://mlsanity.com/paper/2211.07600/) (2022) - [Semi-Supervised Learning with Deep Generative Models](https://mlsanity.com/paper/1406.5298/) (2014) - [GaussianEditor: Swift and Controllable 3D Editing with Gaussian Splatting](https://mlsanity.com/paper/2311.14521/) (2023) - [Text2Mesh: Text-Driven Neural Stylization for Meshes](https://mlsanity.com/paper/2112.03221/) (2021) - [LucidDreamer: Towards High-Fidelity Text-to-3D Generation via Interval Score Matching](https://mlsanity.com/paper/2311.11284/) (2023) - [GaussCtrl: Multi-View Consistent Text-Driven 3D Gaussian Splatting Editing](https://mlsanity.com/paper/2403.08733/) (2024) - [FocalDreamer: Text-driven 3D Editing via Focal-fusion Assembly](https://mlsanity.com/paper/2308.10608/) (2023) - [DreamScene: 3D Gaussian-based Text-to-3D Scene Generation via Formation Pattern Sampling](https://mlsanity.com/paper/2404.03575/) (2024) - [DreamControl: Control-Based Text-to-3D Generation with 3D Self-Prior](https://mlsanity.com/paper/2312.06439/) (2023) - [Debiasing Scores and Prompts of 2D Diffusion for View-consistent Text-to-3D Generation](https://mlsanity.com/paper/2303.15413/) (2023) - [Text-to-3D Generation with Bidirectional Diffusion using both 2D and 3D priors](https://mlsanity.com/paper/2312.04963/) (2023) - [Carve3D: Improving Multi-view Reconstruction Consistency for Diffusion Models with RL Finetuning](https://mlsanity.com/paper/2312.13980/) (2023) - [Hyper-3DG: Text-to-3D Gaussian Generation via Hypergraph](https://mlsanity.com/paper/2403.09236/) (2024) - [Customize your NeRF: Adaptive Source Driven 3D Scene Editing via Local-Global Iterative Training](https://mlsanity.com/paper/2312.01663/) (2023) - [Connecting Consistency Distillation to Score Distillation for Text-to-3D Generation](https://mlsanity.com/paper/2407.13584/) (2024) - [StableDreamer: Taming Noisy Score Distillation Sampling for Text-to-3D](https://mlsanity.com/paper/2312.02189/) (2023) - [Micro-macro Wavelet-based Gaussian Splatting for 3D Reconstruction from Unconstrained Images](https://mlsanity.com/paper/2501.14231/) (2025) ## Built on by (most cited first, 1 of 1) - [CAdam: Context-Adaptive Moment Estimation for 3D Gaussian Densification in Generative Distillation](https://mlsanity.com/paper/2605.20872/) (2026) ## Abstract (arXiv) Versatile 3D tasks (e.g., generation or editing) that distill from Text-to-Image (T2I) diffusion models have attracted significant research interest for not relying on extensive 3D training data. However, T2I models exhibit limitations resulting from prior view bias, which produces conflicting appearances between different views of an object. This bias causes subject-words to preferentially activate prior view features during cross-attention (CA) computation, regardless of the target view condition. To overcome this limitation, we conduct a comprehensive mathematical analysis to reveal the root cause of the prior view bias in T2I models. Moreover, we find different UNet layers show different effects of prior view in CA. Therefore, we propose a novel framework, TD-Attn, which addresses multi-view inconsistency via two key components: (1) the 3D-Aware Attention Guidance Module (3D-AAG) constructs a view-consistent 3D attention Gaussian for subject-words to enforce spatial consistency across attention-focused regions, thereby compensating for the limited spatial information in 2D individual view CA maps; (2) the Hierarchical Attention Modulation Module (HAM) utilizes a Semantic Guidance Tree (SGT) to direct the Semantic Response Profiler (SRP) in localizing and modulating CA layers that are highly responsive to view conditions, where the enhanced CA maps further support the construction of more consistent 3D attention Gaussians. Notably, HAM facilitates semantic-specific interventions, enabling controllable and precise 3D editing. Extensive experiments firmly establish that TD-Attn has the potential to serve as a universal plugin, significantly enhancing multi-view consistency across 3D tasks.