# 3DEditSafe: Defending 3D Editing Pipelines from Unsafe Generation

arXiv: https://arxiv.org/abs/2605.15398 · HTML: https://mlsanity.com/paper/2605.15398/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 28 of its 36 references that are in the map.
- First posted on arXiv 14 May 2026.

## Builds on (most cited first, 28 of 28)

- [Microsoft COCO: Common Objects in Context](https://mlsanity.com/paper/1405.0312/) (2014)
- [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)
- [Mip-NeRF 360: Unbounded Anti-Aliased Neural Radiance Fields](https://mlsanity.com/paper/2111.12077/) (2021)
- [Safe Latent Diffusion: Mitigating Inappropriate Degeneration in Diffusion Models](https://mlsanity.com/paper/2211.05105/) (2022)
- [BlendedMVS: A Large-scale Dataset for Generalized Multi-view Stereo Networks](https://mlsanity.com/paper/1911.10127/) (2019)
- [Instruct-NeRF2NeRF: Editing 3D Scenes with Instructions](https://mlsanity.com/paper/2303.12789/) (2023)
- [Gaussian Grouping: Segment and Edit Anything in 3D Scenes](https://mlsanity.com/paper/2312.00732/) (2023)
- [Diffusion Model-Based Image Editing: A Survey](https://mlsanity.com/paper/2402.17525/) (2024)
- [SneakyPrompt: Jailbreaking Text-to-image Generative Models](https://mlsanity.com/paper/2305.12082/) (2023)
- [Unsafe Diffusion: On the Generation of Unsafe Images and Hateful Memes From Text-To-Image Models](https://mlsanity.com/paper/2305.13873/) (2023)
- [Self-correcting LLM-controlled Diffusion Models](https://mlsanity.com/paper/2311.16090/) (2023)
- [EditSplat: Multi-View Fusion and Attention-Guided Optimization for View-Consistent 3D Scene Editing with 3D Gaussian Splatting](https://mlsanity.com/paper/2412.11520/) (2024)
- [Universal Prompt Optimizer for Safe Text-to-Image Generation](https://mlsanity.com/paper/2402.10882/) (2024)
- [3DitScene: Editing Any Scene via Language-guided Disentangled Gaussian Splatting](https://mlsanity.com/paper/2405.18424/) (2024)
- [A Survey on 3D Gaussian Splatting Applications: Segmentation, Editing, and Generation](https://mlsanity.com/paper/2508.09977/) (2025)
- [PromptGuard: Soft Prompt-Guided Unsafe Content Moderation for Text-to-Image Models](https://mlsanity.com/paper/2501.03544/) (2025)
- [Side Effects of Erasing Concepts from Diffusion Models](https://mlsanity.com/paper/2508.15124/) (2025)
- [CTRL-D: Controllable Dynamic 3D Scene Editing with Personalized 2D Diffusion](https://mlsanity.com/paper/2412.01792/) (2024)
- [3D-LATTE: Latent Space 3D Editing from Textual Instructions](https://mlsanity.com/paper/2509.00269/) (2025)
- [AdLift: Lifting Adversarial Perturbations to Safeguard 3D Gaussian Splatting Assets Against Instruction-Driven Editing](https://mlsanity.com/paper/2512.07247/) (2025)
- [ShieldDiff: Suppressing Sexual Content Generation from Diffusion Models through Reinforcement Learning](https://mlsanity.com/paper/2410.05309/) (2024)
- [NSFW-Classifier Guided Prompt Sanitization for Safe Text-to-Image Generation](https://mlsanity.com/paper/2506.18325/) (2025)
- [Splats in Splats: Robust and Effective 3D Steganography towards Gaussian Splatting](https://mlsanity.com/paper/2412.03121/) (2024)
- [Safety Without Semantic Disruptions: Editing-free Safe Image Generation via Context-preserving Dual Latent Reconstruction](https://mlsanity.com/paper/2411.13982/) (2024)
- [Comprehensive Evaluation and Analysis for NSFW Concept Erasure in Text-to-Image Diffusion Models](https://mlsanity.com/paper/2505.15450/) (2025)
- [Prompt-Based Safety Guidance Is Ineffective for Unlearned Text-to-Image Diffusion Models](https://mlsanity.com/paper/2511.04834/) (2025)
- [GIF: A Conditional Multimodal Generative Framework for IR Drop Imaging in Chip Layouts](https://mlsanity.com/paper/2604.09999/) (2026)

## Abstract (arXiv)

Recent advances in 3D generative editing, particularly pipelines based on 3D Gaussian Splatting (3DGS), have achieved high-fidelity, multi-view-consistent scene manipulation from text prompts. However, we find that these pipelines also introduce new safety risks when unsafe prompts produce edits that are propagated and optimized across views. In this work, we study unsafe generation in 3D editing pipelines and show that such behavior can lead to coherent, undesirable Not-Safe-For-Work (NSFW) content in the final 3D representation. To address this, we propose 3DEditSafe, a safety-regularized 3D editing framework that constrains unsafe semantic propagation during optimization. 3DEditSafe combines generation-stage safety guidance with rendered-view 3D safety regularization, safe semantic projection, residue suppression, and mask-aware preservation to steer optimization away from unsafe editing directions. We evaluate our approach on EditSplat scenes using an object-compatible unsafe prompt benchmark and show that 2D safety guidance alone is not consistently sufficient to prevent unsafe 3D edits. 3DEditSafe reduces unsafe semantic alignment and view-level attack success rates, while revealing a safety-quality tradeoff in which stronger unsafe suppression can introduce artifacts or reduce unsafe-prompt fidelity. To our knowledge, this work is the first attempt to study and defend against unsafe generation in text-driven 3D editing pipelines, highlighting the need for safety mechanisms that operate directly on optimized 3D representations.