Cited by 220Generative Adversarial Networks and Image SynthesisAdversarial Robustness in Machine LearningEthics and Social Impacts of AI

The DOI matches its Crossref record; 43 of the 44 references it cites were found in a registry or an index.

Safe Latent Diffusion: Mitigating Inappropriate Degeneration in Diffusion Models

Open access

Schramowski, Patrick, Brack, Manuel, Deiseroth, Björn, Kersting, Kristian·2023·2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)

DOI 10.1109/cvpr52729.2023.02157

Open this paper in CoChat — free account Open PDF

Abstract

Text-conditioned image generation models have recently achieved astonishing results in image quality and text alignment and are consequently employed in a fast-growing number of applications. Since they are highly data-driven, relying on billion-sized datasets randomly scraped from the internet, they also suffer, as we demonstrate, from degenerated and biased human behavior. In turn, they may even reinforce such biases. To help combat these undesired side effects, we present safe latent diffusion (SLD). Specifically, to measure the inappropriate degeneration due to unfiltered and imbalanced training sets, we establish a novel image generation test bed-inappropriate image prompts (I2P)-containing dedicated, real-world image-to-text prompts covering concepts such as nudity and violence. As our exhaustive empirical evaluation demonstrates, the introduced SLD removes and suppresses inappropriate image parts during the diffusion process, with no additional training required and no adverse effect on overall image quality or text alignment.11Code available at https://huggingface.co/docs/diffusers/api/pipelines/stable.diffusion.safe

Integrity signals

Registry

Crossref · 10.1109/cvpr52729.2023.02157

Cites

43 of 43 references verified0 citation errors · 0 not checkable

Statistics

Recomputing p-values needs the paper’s full text.

Claims the paper makes

We couldn’t retrieve this paper’s full text, so its claims and self-check weren’t run. The registry and citation checks still apply.

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

Language Models are Few-Shot Learners
Brown, T. B., Mann, B., Ryder, N., Subbiah, M., Kaplan, J., Dhariwal, P., Neelakantan, A., Shyam, P., Sastry, G., Askell, A., et al. (2020) · arXiv · doi:10.48550/arxiv.2005.14165 · Cited by 23,372
Learning Transferable Visual Models From Natural Language Supervision
Radford, A., Kim, J. W., Hallacy, C., Ramesh, A., Goh, G., Agarwal, S., Sastry, G., Askell, A., Mishkin, P., Clark, J., et al. (2021) · arXiv · doi:10.48550/arxiv.2103.00020 · Cited by 15,293
Denoising Diffusion Probabilistic Models
Ho, J., Jain, A., & Abbeel, P. (2020) · arXiv · doi:10.48550/arxiv.2006.11239 · Cited by 8,712
On the Dangers of Stochastic Parrots: Can Language Models Be Too Big? 🦜
Bender, E. M., Gebru, T., McMillan-Major, A., & Shmitchell, S. (2021) · Proceedings of the 2021 ACM Conference on Fairness, Accountability, and Transparency · doi:10.1145/3442188.3445922 · Cited by 6,988 · cited 1×
“These concerns include, for instance, models producing stereotypical and derogatory content [3] and gender and racial biases [10, 24, 38, 41].”
Background
Semantics derived automatically from language corpora contain human-like biases
Caliskan, A., Bryson, J. J., & Narayanan, A. (2017) · Science · doi:10.1126/science.aal4230 · Cited by 2,887
Photorealistic Text-to-Image Diffusion Models with Deep Language Understanding
Saharia, C., Chan, W., Saxena, S., Li, L., Whang, J., Denton, E., Ghasemipour, S. K. S., Ayan, B. K., Mahdavi, S. S., Lopes, R. G., et al. (2022) · arXiv · doi:10.48550/arxiv.2205.11487 · Cited by 2,528
Hierarchical Text-Conditional Image Generation with CLIP Latents
Ramesh, A., Dhariwal, P., Nichol, A., Chu, C., & Chen, M. (2022) · arXiv · doi:10.48550/arxiv.2204.06125 · Cited by 2,511 · cited 5×
“Inspired by existing content policies of deployed text-to-image DMs such as DALL-E, the following text prompt is used to define a general inappropriate safety concept: “hate, harassment, violence, suffering, humiliation, harm, suicide, sexual, nudity, bodily fluids, blood, obscene gestures, illegal activity, drug use,…”
BackgroundInfluential
Datasheets for Datasets
Gebru, T., Morgenstern, J., Vecchione, B., Vaughan, J. W., Wallach, H., Daumé, H., & Crawford, K. (2018) · arXiv · doi:10.48550/arxiv.1803.09010 · Cited by 2,474
Datasheets for datasets
Gebru, T., Morgenstern, J., Vecchione, B., Vaughan, J. W., Wallach, H., III, H. D., & Crawford, K. (2021) · Communications of the ACM · doi:10.1145/3458723 · Cited by 2,469
Prompt-to-Prompt Image Editing with Cross Attention ControlPreprint
Hertz, A., Mokady, R., Tenenbaum, J., Aberman, K., Pritch, Y., & Cohen‐Or, D. (2022) · arXiv (Cornell University) · doi:10.48550/arxiv.2208.01626 · Cited by 1,894

Flagged first, then by citations

Show all 44 references

1 of the 44 references couldn’t be matched to a known paper.

Research works printed but not linked: 2

birhane, 2021, Multimodal datasets: misogyny, pornography, andma-lignant stereotypes, CoRR, abs 2110 1963
valevski, 0, Unitune Text-driven image editing by fine tuning an image generation model on a single image

Sorted by citations

Items 101–200 of 220

The Safety Illusion? Testing the Boundaries of Concept Removal in Diffusion Models
Pan, Y., Luo, T., Li, Y., Xing, W., Chen, M., & Han, M. (2025) · IEEE Transactions on Image Processing · doi:10.1109/tip.2025.3620665 · Cited by 1
Universally Unfiltered and Unseen: Input-Agnostic Multimodal Jailbreaks against Text-to-Image Model Safeguards
Yan, S., Wei, H., Fei, J., Yang, G., Zhao, Z., & Wang, Z. (2025) · ACM International Conference on Multimedia (ACM MM) · doi:10.1145/3746027.3754928 · Cited by 1
FORGET ME: Federated Unlearning for Face Generation Models
Qi, F., Liu, A., Zhang, Z., & Xu, C. (2025) · ACM International Conference on Multimedia (ACM MM) · doi:10.1145/3746027.3754935 · Cited by 3
Mitigating Stereotypes in Text-to-Image Generation: A Novel Perspective of Selective Neural Suppression
Zhou, J., Gao, J., Guo, X., Wu, H., Liu, Q., Zhao, X., Wei, H., Yao, X., & Wei, X. (2025) · ACM International Conference on Multimedia (ACM MM) · doi:10.1145/3746027.3755293 · Cited by 1
Corer: Concept Residue Erasing in Text-to-Image Diffusion Models
Liu, Y., An, J., Chen, H., Zhang, W., Li, M., Wu, D., Gu, J., Lin, Z., & Wang, W. (2025) · IEEE International Conference on Multimedia and Expo (ICME) · doi:10.1109/icme59968.2025.11210155 · Cited by 1
Antelope: Potent and Concealed Jailbreak Attack Strategy
Zhao, X., Chen, X., & Gao, H. (2025) · ACM International Conference on Information and Knowledge Management (CIKM) · doi:10.1145/3746252.3761283 · Cited by 3
Don’t Let Your Robot Be Harmful: Responsible Robotic Manipulation via Safety-As-Policy
Ni, M., Lei Zhang, A., Chen, Z., Bai, K., Chen, Z., Zhang, J., Lei Zhang, A., & Zuo, W. (2025) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2025.3623027 · Cited by 1
Concept Corrector: Erase Concepts on the Fly for Text-to-Image Diffusion Models
Meng, Z., Peng, B., Jin, X., Lyu, Y., Wang, W., Dong, J., & Tan, T. (2026) · Lecture notes in computer science · doi:10.1007/978-981-95-5567-3_7 · Cited by 1
Inference Time Debiasing Concepts in Diffusion Models
Kupssinskü, L. S., Bochernitsan, M. N., Kopper, J., Parraga, O., & Barros, R. C. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-15984-7_6 · Cited by 1
How to Train Your Text-to-Image Model: Evaluating Design Choices for Synthetic Training Captions
Brack, M., Katakol, S., Friedrich, F., Schramowski, P., Ravi, H., Kersting, K., & Kale, A. (2025) · 2025 IEEE/CVF International Conference on Computer Vision Workshops (ICCVW) · doi:10.1109/iccvw69036.2025.00711 · Cited by 5
Self-citation
A Comparative Survey of Social Bias in Text and Image Generation: Gaps, Directions and Compliance with the EU AI Act
Rad, A.-C., & Lemnaru, C. (2025) · International Conference on Computational Photography · doi:10.1109/iccp68926.2025.11427185 · Cited by 1
Sculpting Memory: Multi-Concept Forgetting in Diffusion Models via Dynamic Mask and Concept-Aware Optimization
Li, G., Xiao, Y., Ji, J., Deng, K., Hui, B., Guo, L., & Ma, X. (2025) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51701.2025.01828 · Cited by 1
AutoPrompt: Automated Red-Teaming of Text-to-Image Models via LLM-Driven Adversarial Prompts
Liu, Y., Zhang, W., Chen, H., Wang, L., Jia, X., Lin, Z., & Wang, W. (2025) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51701.2025.01631 · Cited by 1
TRCE: Towards Reliable Malicious Concept Erasure in Text-to-Image Diffusion Models
Chen, R., Guo, H., Wang, L., Zhang, C. T., Nie, W., & Liu, A.-A. (2025) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51701.2025.01759 · Cited by 1
Selective subspace unlearning for text to image diffusion models
Dubey, A., Sharma, M., & Kancharla, P. (2026) · Pattern Recognition Letters · doi:10.1016/j.patrec.2026.07.006 · Cited by 1
VA3: Virtually Assured Amplification Attack on Probabilistic Copyright Protection for Text-to-Image Generative Models
Li, X., Shen, Q., & Kawaguchi, K. (2024) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr52733.2024.01175 · Cited by 6
MultiDiffEditAttack: A Multi-Modal Black-Box Jailbreak Attack on Image Editing Models
Chen, P., Chen, F., & Guo, L. (2025) · Electronics · doi:10.3390/electronics14050899 · Cited by 1
Machine Unlearning for Foundation Models
Chen, P., & Liu, S. (2025) · Springer eBooks · doi:10.1007/978-3-031-76770-8_11 · Cited by 1
Prompt Engineering for Safety Red-Teaming: A Case Study on Text-to-Image Diffusion Models
Chen, P., & Liu, S. (2025) · Springer eBooks · doi:10.1007/978-3-031-76770-8_18 · Cited by 1
S 3 Diffuser: Frequency selected state space guided diffusion model for multimodal fusion classification
Zhao, J., Jiao, L., Li, L., Ma, M., Liu, X., Wang, C., Liu, F., Ma, W., & Yang, S. (2025) · Information Fusion · doi:10.1016/j.inffus.2025.103447 · Cited by 1
ICE: Intercede Concept Erasure in Text-to-Image Diffusion Models
Lin, Y., Huang, N., Huang, K., Liu, H. M., Yan, Y., Guo, J. M., Lee, T., & Li, X. (2025) · ACM International Conference on Multimedia (ACM MM) · doi:10.1145/3746027.3754992 · Cited by 1
Netdiffus: Network Traffic Generation by Diffusion Models Through Time-Series ImagingPreprint
Sivaroopan, N., Bandara, D., Madarasingha, C., Jourjon, G., Jayasumana, A. P., & Thilakarathna, K. (2024) · SSRN Electronic Journal · doi:10.2139/ssrn.4772614 · Cited by 1
SINTESIS TEKS KE GAMBAR: TINJAUAN ATAS DATASET
Arifin, S. R. (2024) · Jurnal EEICT (Electric Electronic Instrumentation Control Telecommunication) · doi:10.31602/eeict.v7i1.13066 · Cited by 1
Lesiondiff: enhanced breast cancer classification via dynamic lesion amplification using diffusion models
Lai, Y., Liu, Y., Zhang, Q., Shang, J., Qiu, X., & Yan, J. (2024) · Computer Methods in Biomechanics and Biomedical Engineering Imaging & Visualization · doi:10.1080/21681163.2024.2433478
PRJ: Perception–Retrieval–Judgement for Generated Images
Fu, Q., Jing, Z., Ying, Z., & Li, X. (2025) · Electronics · doi:10.3390/electronics14122354
Adversarial Paraphrase Substitution: Jailbreak Attacks on Text-to-Image Diffusion Models
Zhao, B., Lu, B., Gao, X., & Shi, M. (2025) · International Seminar on Artificial Intelligence, Networking and Information Technology (AINIT) · doi:10.1109/ainit65432.2025.11035508
Generative AI Models for Images with Copyright-Free TrainingPreprint
Xu, T., Cohen, M. C., & Clark, J. J. (2025) · SSRN Electronic Journal · doi:10.2139/ssrn.5331280
TFCM: Tuning-Free Facial Concept-Erasure in Text-to-Image Models Through Attention and Sample Modulation
Guo, Q., Luo, G., Bai, Z., & Zhu, Y. (2025) · Lecture notes in computer science · doi:10.1007/978-981-96-6594-5_13
Mitigating Replication and Bias in Diffusion-Based Face Image Synthesis Models With Rewards
Perera, M. V., & Patel, V. M. (2025) · IEEE Transactions on Biometrics Behavior and Identity Science · doi:10.1109/tbiom.2025.3590344
VSF: Simple, Efficient, and Effective Negative Guidance in Few-Step Image Generation Models By Value Sign FlipPreprint
Guo, W., & Du, S. (2025) · Qeios · doi:10.32388/poy29b
Guided Safe Diffusion: Prohibiting Diffusion Models from Generating Inappropriate Content
Jiang, S., Zhang, R., Yang, X., Dong, B., & Huang, K. (2025) · Communications in computer and information science · doi:10.1007/978-981-96-7036-9_11
Concept Replacer: Replacing Sensitive Concepts in Diffusion Models via Precision Localization
Zhang, L., Xie, Y., Fu, Y., & Chen, P. (2025) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr52734.2025.00765
Is Perturbation-Based Image Protection Disruptive to Image Editing?
Tang, Q., Ayambem, B., Chuah, M. C., & Bharati, A. (2025) · Proceedings - International Conference on Image Processing · doi:10.1109/icip55913.2025.11084494
Mitigating inappropriate concepts in text-to-image generation with attention-guided Image editing
Oh, J., Jeong, J.-Y., Hong, Y.-G., & Jeong, J.-W. (2025) · PeerJ Computer Science · doi:10.7717/peerj-cs.3170
CSS: Clustering-guided SliderSpace for Under-represented Visual Concept Exploration of Diffusion Models
Chen, J.-C. (2025) · IEEE International Conference on Advanced Video and Signal Based Surveillance (AVSS) · doi:10.1109/avss65446.2025.11149711
AllGuard: A Multimodal Large Language Model for Edge-Deployed Content Security Assessment
Qu, W., Lu, W., Chen, C., Li, T., Chen, H.-D., Du, R.-Q., & Ling, X.-Y. (2025) · Journal of Computer Science and Technology · doi:10.1007/s11390-025-5508-7
ACE: Concept Editing in Diffusion Models without Performance Degradation
Wang, R., Fang, J., Li, J., Chen, H., Shi, J., Wang, K., & Wang, X. (2025) · ACM International Conference on Multimedia (ACM MM) · doi:10.1145/3746027.3755749 · Cited by 5
SafeCFG: Controlling Harmful Features with Dynamic Safe Guidance for Safe Generation
Pan, J., Li, L., Gao, H., Zha, Z.-J., Huang, Q., & Luo, J. (2025) · ACM International Conference on Multimedia (ACM MM) · doi:10.1145/3746027.3754717 · Cited by 1
Let’s Talk AI with Computer Science Expert Daniel Neider
Neider, D., & Steffen, B. (2025) · Lecture notes in computer science · doi:10.1007/978-3-032-09008-9_14
Culture-based Adversarial Attack on Text-to-Image Models
Yang, F., Zhang, C., & Wang, L. (2025) · IEEE International Conference on Multimedia and Expo (ICME) · doi:10.1109/icme59968.2025.11209989
Steering LLMs Towards Safer Shores
Batenburg, V., Wijnholds, G., & Gadyatskaya, O. (2025) · Communications in computer and information science · doi:10.1007/978-981-95-4409-7_4
Securing GenAI Deployments and Preventing Misuse
Sikos, L. F. (2025) · SpringerBriefs in cybersecurity · doi:10.1007/978-3-032-05250-6_4
Towards SFW sampling for diffusion models via external conditioning
Reyes, C. C., Fontbona, J., & Tobar, F. A. (2025) · International Joint Conference on Neural Networks (IJCNN) · doi:10.1109/ijcnn64981.2025.11227152
What Lurks Within? Concept Auditing for Shared Diffusion Models at Scale
Yuan, X., Ma, X., Guo, L., & Zhang, L. (2025) · ACM SIGSAC Conference on Computer and Communications Security (CCS) · doi:10.1145/3719027.3765052
Safety Assessment of 3D Generation Models in AR/VR Applications
Tang, X., Ma, W., Bao, Y., Xue, M., Wen, S., & Xiang, Y. (2025) · doi:10.1145/3733800.3763268
Compliance Rating Scheme: A Data Provenance Framework for Generative AI Datasets
Boháček, M., & Echávarri, I. V. (2025) · ACM International Conference on Multimedia (ACM MM) · doi:10.1145/3746027.3758141 · Cited by 1
CGCE: Classifier-Guided Concept Erasure in Generative Models
Nguyen, V., & Patel, V. M. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-37429-5_30 · Cited by 3
MAC: A MultiModal Automatic Attack Framework for Text-to-Image Models Optimized for Chinese
Zhang, Y. (2025) · International Conference on Image Processing, Computer Vision and Machine Learning (ICICML) · doi:10.1109/icicml67980.2025.11333461
Deep Concept Forgetting in Text-to-Image Diffusion Models
Zhou, Y., Zheng, D., Wu, X., & Zheng, W. (2026) · Lecture notes in computer science · doi:10.1007/978-981-95-5693-9_4
Dual-path consistency constrained concept erasure for text-to-image diffusion models
Bai, X., Song, D., Sun, P., & Yue, S. (2026) · Multimedia Systems · doi:10.1007/s00530-025-02168-8 · Cited by 1
FP-Director: Direction-Guided Latent Code Refinement for Facial-Preference Alignment in Text-to-Image Diffusion
Jiang, Y., Lyu, Y., Ma, T., Peng, B., & Dong, J. (2026) · Lecture notes in computer science · doi:10.1007/978-981-95-6123-0_22
Transcending Image Generation Paradigms: Fusion of Transformer-Based Latent Diffusion and Semantic Clip Embeddings
Mandapati, S. V. R., & Mandapati, V. M. N. R. (2025) · International Conference on Computing, Semiconductor, Mechatronics, Intelligent Systems and Communications (COSMIC) · doi:10.1109/cosmic67569.2025.11380849
Guardians of Generation: Dynamic Inference-Time Copyright Shielding with Adaptive Guidance for AI Image Generation
Roy, S., Mishra, A., Sharma, A. S., Karande, S. S., & Mandal, M. (2025) · IEEE International Conference on Computer Vision Workshops · doi:10.1109/iccvw69036.2025.00268
Safety Without Semantic Disruptions: Editing-free Safe Image Generation via Context-preserving Dual Latent Reconstruction
Vice, J., Akhtar, N., Shah, M., Hartley, R. I., & Mian, A. S. (2025) · IEEE International Conference on Computer Vision Workshops · doi:10.1109/iccvw69036.2025.00243
Movable-type eraser: separable, recoverable, and sustainable multi-concept eraser in diffusion models
Zhao, M., Zhang, L., & Yin, B. (2026) · Scientia Sinica Informationis · doi:10.1360/ssi-2025-0204
Debias-DPO: Debiasing Diffusion-based Face Image Generation with Direct Preference Optimization
Perera, M. V., & Patel, V. M. (2025) · IEEE International Joint Conference on Biometrics (IJCB) · doi:10.1109/ijcb65343.2025.11411388
Date: dynamic attention-based toxicity elimination mechanism for safe image generation
Cha, Y.-H., Park, J., & Lee, Y. (2026) · Multimedia Systems · doi:10.1007/s00530-025-02166-w
OrthoEraser: Coupled-Neuron Orthogonal Projection for Concept Erasure
Shi, C., Wu, W., Shen, F., Zhu, X., Hu, K., & Wang, Z. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-37232-1_30 · Cited by 7
Video Identifying and Eraser: Use Multi-Task Cascaded Convolutional Neural Network to Enhance Safety in a Text-to-Video Diffusion Model
Lin, S., Zhou, R., & Wang, Y. (2026) · Applied Sciences · doi:10.3390/app16062995
PromptNeedling: Jailbreaking Text-to-Video Generative Models
Mu, Y., Li, H., Zhang, R., Jiang, W., & Fan, W. (2025) · Globecom · doi:10.1109/globecom59602.2025.11432140
IDENTITYGUARD: Context-Aware Restriction and Provenance for Personalized Synthesis
Zhang, L., Xie, Y., & Chen, P. (2026) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp55912.2026.11463895
Training-Free Framework For Defending Unsafe Image Synthesis Attack
Park, J., Hwang, J., Ryu, I., Park, H. Y., Kim, J., & Lee, J.-S. (2026) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp55912.2026.11461295
NeurErase: Selective Deactivation of Neurons for Erasing Concepts in Diffusion Models
Chen, Z., Lyu, Y., Meng, Z., Rao, H., Li, N., Wang, S., & Shan, C. (2026) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp55912.2026.11464730
Automated Red Teaming for Text-to-Image Models Through Feedback-Guided Prompt Iteration with Vision-Language Models
Xu, W., Chen, K., Qiu, J., Zhang, Y.-Y., Wang, R., Mao, J., Zhang, T.-W., & Wang, L. (2025) · IEEE International Conference on Computer Vision · doi:10.1109/iccv51701.2025.01726 · Cited by 4
DSA: A Dual-Safety Approach for Test-Time Temporal Risk Removal of Video Generation
Liu, Q., Liu, K., Liu, X., Gu, X., Ren, Z., Zhang, Y., & Liu, W. (2026) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2026.3689683 · Cited by 1
JailbreakDiffBench: A Comprehensive Benchmark for Jailbreaking Diffusion Models
Jin, X., Weng, Z., Guo, H., Yin, C., Cheng, S., Shen, G., & Zhang, X. (2025) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51701.2025.01528
ZIUM: Zero-Shot Intent-Aware Adversarial Attack on Unlearned Models
Yook, H. J., Jhun, G. S., Cho, J. H., Jeon, M., Kim, D., Kim, T. H., & Lee, Y. K. (2025) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51701.2025.00374
PLA: Prompt Learning Attack Against Text-To-Image Generative Models
Lyu, X., Liu, Y., Li, Y., & Xiao, B. (2025) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51701.2025.01565
Safe Vision-Language Models via Unsafe Weights Manipulation
Moreno D’Incà, Peruzzo, E., Xu, X., Shi, H., Sebe, N., & Mancini, M. (2026) · IEEE Winter Conference on Applications of Computer Vision · doi:10.1109/wacv61042.2026.00455 · Cited by 3
Unveiling the Risk of Unsafe Image Generation in Stable Diffusion Through a Cross-Attention Mechanism
Zhuang, Y., Jing, Y., Yi, W., Xu, X., & Wang, J. (2026) · Future Internet · doi:10.3390/fi18050248
Buster: Implanting Semantic Backdoor Into Text Encoder to Mitigate NSFW Content Generation
Zhao, X., Chen, X., Xuan, Y., Zhao, Z., Li, X. Q., Jia, X., & Wang, X. (2026) · Lecture notes in computer science · doi:10.1007/978-981-92-0375-8_27 · Cited by 3
AICE: Anchor Images Concept Erasure in Text-to-Image Diffusion Models
Zhang, Z., Lu, H., Zhou, G., & He, T. (2026) · International Conference on Computer, Artificial Intelligence and Control Engineering (CAICE) · doi:10.1145/3804601.3804691
Machine Unlearning for AI Safety
Chen, Y., & Liu, S. (2026) · Synthesis lectures on computer vision · doi:10.1007/978-3-032-17282-2_13
Unlearning Through Internal Model Analysis: Sparsity, Saliency, and Attribution
Jia, J. (2026) · Synthesis lectures on computer vision · doi:10.1007/978-3-032-17282-2_9
Data Integrity for Machine Unlearning
Wang, C., & Zhang, Y. (2026) · Synthesis lectures on computer vision · doi:10.1007/978-3-032-17282-2_6
User Perceptions of Emotions in the Facial Expressions of AI-Generated Characters
Kim, S., & Yu, H.-J. (2026) · Journal of Digital Contents Society · doi:10.9728/dcs.2026.27.4.1037
MPDA: Multimodal Prompt Decoupling Attack on the Safety Filters in Text-to-Image Models
Peng, X., Jiang, J., Meng, T., Li, S., Zhang, W., Yu, N., & Chen, K. X. (2026) · IEEE Transactions on Dependable and Secure Computing · doi:10.1109/tdsc.2026.3696145
Advancing Image Generation: Integrating Transformer-Based Latent Diffusion With Semantic Clip Embeddings
Sriram, S., Karthikeyan, & Nivethitha, V. (2026) · International Conference on Intelligent Computing, Networks, and Security (IC-ICNS) · doi:10.1109/ic-icns68863.2026.11537489
Optimizing Trust and Safety Regions for Text-to-Image Generation in High-Dimensional Manifold Spaces
Yang, X., Li, X., Wang, Y., & Liu, N. (2026) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2026.3699945
When Erasure Fails: Benchmarking Content Removal in Diffusion Models via Vocabulary Substitution Attack
Ruan, N., Fu, C., Huang, T., & Mei, Z. (2026) · IEEE Transactions on Dependable and Secure Computing · doi:10.1109/tdsc.2026.3707740
Bridging the Scale Gap: Augmenting Human Red-Teaming to Uncover Latent Risks in T2I Models
Quaye, J. A., Parrish, A., Rastogi, C., Kahng, M., Inel, O., Aroyo, L. M., & Reddi, V. J. (2026) · ACM Conference on Fairness, Accountability, and Transparency (FAccT) · doi:10.1145/3805689.3812210
Obliviate: Erasing Concepts from Autoregressive Image Generation Models
Shakibania, H., Grebe, J. H., Braun, T., Aktemur, E., Aslani, S., Yiğit, M. G., & Rohrbach, M. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-37324-3_36
The Illusion of High Utility in Safety Alignment of Text-to-Image Diffusion Models
Yousaf, A., Ghosh, S. K., Beetham, J., Bedi, A. S., & Shah, M. K. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-37624-4_17
Latent Space Probing for Adult Content Detection in Video Generative Models
Khatri, A., & Prabhu, C. (2026) · International Conference on Dependable Systems and Networks workshops · doi:10.1109/dsn-w70714.2026.00039
Text-to-Unlearn: Robust Concept Removal in GANs via Text Prompts
Nagasubramaniam, P., Karamchandani, N., Wu, C., & Zhu, S. (2026) · Journal of Cybersecurity and Privacy · doi:10.3390/jcp6040121
Audio Immunization Against Harmful Audio Editing with Diffusion Models
Yu, J., Zhang, H., Zhao, Y., Su, J., & Wei, Y. (2026) · Lecture notes in computer science · doi:10.1007/978-981-92-2859-1_23
MoE-Erase: Mixture-of-Experts with LoRA Adaptation and DPO Routing for Category-Specific Concept Erasure in Diffusion Models
Liu, X. (2026) · Lecture notes in computer science · doi:10.1007/978-981-92-3492-9_28
UniNDM: A Unified Noise-Driven Detection and Mitigation Framework Against Sexual Content in Text-to-Image Generation
Huang, Y., Sun, Y., Chen, H., Zhang, R., Ruan, S., Duan, R., Yuan, M., Dong, Y., Xue, H., Cao, X., et al. (2026) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2026.3713771
Retention-Consistent Concept Unlearning for Diffusion Models
Qin, L., Zhu, T., Wang, L., & Zhou, W. Y. (2026) · Lecture notes in computer science · doi:10.1007/978-981-92-3012-9_13
Forgotten Horizons in Concept Erasure: Safeguarding Close-Proximity Concepts in Text-to-Image Models
Li, J. F., Liu, Y., Yu, R., Zhang, C., Qi, G., Bi, S., Liu, F., Geng, J., & Karray, F. O. (2026) · IEEE Transactions on Information Forensics and Security · doi:10.1109/tifs.2026.3714133
EvoJail: Jailbreaking Text-to-Image Models via Semantic Evolution
Xu, Z., Li, Z. D., Liu, Y., Ma, X. C., Zhang, Y., & Chen, Y. (2026) · Lecture notes in computer science · doi:10.1007/978-981-92-3423-3_33
Safe Image Generation via Lightweight Concept Erasure in Diffusion Models
Geng, X., Hui, S., Wang, Y., Zhou, J. T., & Wang, Z. (2026) · IEEE Transactions on Image Processing · doi:10.1109/tip.2026.3712345
GRAD: Gradient-based Robustness Audit for Multimodal Diffusion Models
He, Z., Wang, Y., Pan, Z., & Ying, Z. (2026) · International Conference on Big Data and Privacy Computing (BDPC) · doi:10.1109/bdpc69980.2026.11608049
MUKSB: Machine unlearning via Kalai-Smorodinsky bargaining
Dubey, A., Sharma, M., & Kancharla, P. (2026) · Expert Systems with Applications · doi:10.1016/j.eswa.2026.133675
Retrieval-Augmented Concept Erasure in Diffusion Models
Yin, Q., Li, H., Zhang, T., & Zhang, Y. (2026) · International Conference on Image and Graphics Processing (ICIGP) · doi:10.1109/icigp68997.2026.11619857
To Erase, or Not to Erase: Robust Training-Free Concept Erasure with Preservation Aware Adaptive Ranked Subspace Expansion
Saha, S., Anguluri, R., & Gaur, M. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-37556-8_33 · Cited by 3
Efficient Concept Unlearning in Latent Diffusion Models via Feature Caching
Sharma, M., Hase, A., & Kancharla, P. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-31920-3_37
PECKER: A Precisely Efficient Critical Knowledge Erasure Recipe For Machine Unlearning in Diffusion Models
Ma, Z., Deng, Z., Tang, H., Chen, J., Zheng, Z., Li, Z., & Chuai, Q. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-31920-3_3
Graph Diffusion History Reconstruction via Feasibility-Aware Markov Chain Monte Carlo Estimation
Zuo, Y., Qiu, R., Chen, L., & Tong, H. (2026) · ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD) · doi:10.1145/3770855.3818096
Patronus: Safeguarding Text-to-Image Models Against Adversarial Fine-Tuning
Li, X., Pang, S., Wu, J., Deng, J., Zhong, H., Chen, Y., Zhang, J., & Xu, W. (2026) · IEEE Transactions on Information Forensics and Security · doi:10.1109/tifs.2026.3723129 · Cited by 1

Authors: Schramowski, P., Brack, M., Deiseroth, B., & Kersting, K.
Affiliations: Technische Universität Darmstadt (DE)
ORCID: Schramowski, Patrick
Brack, Manuel
Kersting, Kristian
Venue: 2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
Publisher: IEEE
Published: 2023-06
Pages: 22522-22531
ISSN: 1063-6919 (ISSN-L)
DOI: 10.1109/cvpr52729.2023.02157
OpenAlex: W4386066536
Semantic Scholar: 0231f2aed9a96cb516242fb57f2cb63f5651c4d8
Cited by: 219
References: 44
Citations by year: 2026: 68 · 2025: 86 · 2024: 55 · 2023: 10
FWCI: 23.31
Type: conference paper
Language: en
Open access: green
arxiv.org
Licence: cc-by
doi.org/10.15223/policy-029 (stm-asf)
doi.org/10.15223/policy-037 (stm-asf)
Locations: IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings (publishedVersion)
arXiv (Cornell University) (submittedVersion, cc-by)
arXiv (Cornell University) (cc-by)
Topics: Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Generative Adversarial Networks and Image Synthesis
Physical Sciences › Computer Science › Artificial Intelligence › Adversarial Robustness in Machine Learning
Physical Sciences › Computer Science › Artificial Intelligence › Ethics and Social Impacts of AI
Keywords: latent diffusion models, text-to-image generation, training data bias, nudity, violence, text alignment

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

Safe Latent Diffusion: Mitigating Inappropriate Degeneration in Diffusion Models

Abstract

Overview

Integrity signals

Verification

Claims the paper makes

Claims other papers make about this one

Citations

Cites

Cited by

Related

Metadata

Record

About these checks