Weeding Out Bad Seeds: Initial-Noise-Robust Unlearning for Text-to-Image Diffusion Models

arXiv:2609.37537 · cs.CV · Submitted 2026-09-29 · Read on arXiv

cs.CV

Submitted: 2026-09-29

Updated: 2026-10-03

Code: https://github.com/black-forest-labs/flux

Terminology

• Diffusion Model — 34× in this paper · explained in 8 episodes
• Diffusion Models — 34× in this paper · explained in 6 episodes
• Accuracy — 27× in this paper · explained in 1 episode
• Robustness — 25× in this paper · explained in 5 episodes
• Ablation — 15× in this paper · explained in 1 episode
• Machine Unlearning — 13× in this paper · explained in 2 episodes
• Benchmark — 8× in this paper · explained in 2 episodes
• Regularization — 7× in this paper · explained in 3 episodes
• Adaptive Sampling — 5× in this paper · explained in 1 episode
• Adaptive Strategy — 5× in this paper · explained in 1 episode
• Ablation study — 4× in this paper · explained in 1 episode
• Adversarial Robustness — 4× in this paper · explained in 3 episodes
• Alignment — 4× in this paper · explained in 1 episode
• Conference on Computer Vision and Pattern Recognition — 4× in this paper
• Computational overhead — 3× in this paper · explained in 2 episodes
• Direct Unlearning Optimization — 3× in this paper
• Transformer — 3× in this paper · explained in 2 episodes
• Unlearning Accuracy — 3× in this paper
• Adam — 2× in this paper · explained in 2 episodes
• Adaptive sampling strategy — 2× in this paper · explained in 1 episode
• Adversarial Attacks — 2× in this paper · explained in 1 episode
• Core — 2× in this paper · explained in 2 episodes
• Generative Model — 2× in this paper · explained in 2 episodes
• Generative Models — 2× in this paper · explained in 2 episodes
• In-domain Retain Accuracy — 2× in this paper
• Latent Diffusion Model — 2× in this paper · explained in 2 episodes
• Latent Diffusion Models — 2× in this paper · explained in 2 episodes
• Objective Function — 2× in this paper · explained in 2 episodes
• Rectified Flow — 2× in this paper · explained in 3 episodes
• Reward Hacking — 2× in this paper · explained in 8 episodes
• Activation Function — 1× in this paper · explained in 1 episode
• Active Regions — 1× in this paper · explained in 1 episode
• Adaptation — 1× in this paper · explained in 1 episode
• Approximation Bound — 1× in this paper · explained in 1 episode
• Attack Success Rate (ASR) — 1× in this paper · explained in 2 episodes
• Benchmarking — 1× in this paper · explained in 3 episodes
• Concept Ablation — 1× in this paper
• Continual Learning — 1× in this paper · explained in 2 episodes
• Contrastive Learning — 1× in this paper · explained in 5 episodes
• Cross-domain Retain Accuracy — 1× in this paper
• Distillation — 1× in this paper · explained in 2 episodes
• Distribution Shift — 1× in this paper · explained in 5 episodes
• Embeddings — 1× in this paper · explained in 2 episodes
• Exposure Bias — 1× in this paper · explained in 2 episodes
• Fidelity — 1× in this paper · explained in 2 episodes
• Generalization — 1× in this paper · explained in 13 episodes
• Graceful Degradation — 1× in this paper · explained in 2 episodes
• Loss Function — 1× in this paper · explained in 2 episodes
• Parameter Efficiency — 1× in this paper · explained in 3 episodes
• Semantic Alignment — 1× in this paper · explained in 3 episodes
• Steering — 1× in this paper · explained in 2 episodes
• Theoretical Guarantees — 1× in this paper · explained in 2 episodes
• Trace — 1× in this paper · explained in 2 episodes

Sources

• Erasing Undesirable Concepts in Diffusion Models with Adversarial Preservation
• Fantastic Targets for Concept Erasure in Diffusion Models and Where To Find Them
• Score Forgetting Distillation: A Swift, Data-Free Method for Machine Unlearning in Diffusion Models
• Prompting4Debugging: Red-Teaming Text-to-Image Diffusion Models by Finding Problematic Prompts
• SalUn: Empowering Machine Unlearning via Gradient-based Weight Saliency in Both Image Classification and Generation
• If at First You Don't Succeed, Try, Try Again: Faithful Diffusion-based Text-to-Image Generation by Selection
• CARINOX: Inference-time Scaling with Category-Aware Reward-based Initial Noise Optimization and Exploration
• Erasure or Erosion? Evaluating Compositional Degradation in Unlearned Text-To-Image Diffusion Models
• All Seeds Are Not Equal: Enhancing Compositional Text-to-Image Generation with Reliable Random Seeds
• Unrolled Generative Adversarial Networks
• GLIDE: Towards Photorealistic Image Generation and Editing with Text-Guided Diffusion Models
• Input Perturbation Reduces Exposure Bias in Diffusion Models
• Red-Teaming the Stable Diffusion Safety Filter
• Six-CD: Benchmarking Concept Removals for Benign Text-to-image Diffusion Models
• Contrastive Learning with Hard Negative Samples
• High-Resolution Image Synthesis with Latent Diffusion Models
• Denoising Diffusion Implicit Models
• Rethinking the Inception Architecture for Computer Vision
• Reward Hacking in the Era of Large Models: Mechanisms, Emergent Misalignment, Challenges
• Diffusion Models: A Comprehensive Survey of Methods and Applications

Related papers

• Loss Knows Best: Detecting Annotation Errors in Videos via Loss Trajectories
• AnchorWeave: World-Consistent Video Generation with Retrieved Local Spatial Memories
• Benchmarking the Robustness of Foundation Models for Mammography under Domain Shift
• MambaX-Net: Dual-Input Mamba-Enhanced Cross-Attention Network for Longitudinal MRI Segmentation
• TeleOCR: Navigating Document Parsing Across Digital and Camera-Captured Documents
• A Survey on Efficient Vision-Language-Action Models