Child Sexual Abuse Material

Follow this topic Ask about this topic

Momentum

0 papers in the last four weeks, down 100% on the four weeks before. 0.0% of all new papers.

Jul 6Week of Sep 21

Latest papers 14

All topics

Weeding Out Bad Seeds: Initial-Noise-Robust Unlearning for Text-to-Image Diffusion Models
Sep 29, 2026Arian Komaei Koma, Seyed Amir Kasaei, Aida Aryafar +6Text-To-Image Diffusion Models Diffusion Models
Sharif University of Technology · Hong Kong University of Science and Technology
ConVAWG: A Retrieval-Grounded Framework for Controlled Synthetic Dialogue Generation in Violence Against Women and Girls
Aug 11, 2026Chen Lyu, Xingwei Tan, Simon Cullen +4Conversational Datasets Child Sexual Abuse Material
University of Warwick, UK · University of Sheffield, UK · Forensic Capability Network, UK +1
Towards an LLM-based method for quantifying the sexual content in song lyrics
Aug 9, 2026Ignacio M. SticcoChild Sexual Abuse Material
Ignacio M. Sticco
Test-Time Scaling for Safe Text-Guided Image Generation via Intermediate Clean Estimates
Aug 4, 2026Jinya Sakurai, Shueicheng Yan, Xun XuText-To-Image Diffusion Models Text-To-Image
1Nanyang Technological University · 2National University of Singapore · Institute of Advanced Intelligence and Computing (IAIC), A*STAR +1
Detecting CSAM Text-to-Image LoRAs From Weights
Jul 28, 2026David Demitri Africa, Cate Heine, Nadine Staes-Polet +1Child Sexual Abuse Material Adversarial Examples
UK AI Security Institute
UniNDM: A Unified Noise-driven Detection and Mitigation Framework Against Sexual Content in Text-to-Image Generation
Jul 18, 2026Yao Huang, Yitong Sun, Huanran Chen +8Text-To-Image Diffusion Models Child Sexual Abuse Material
Institute of Artificial Intelligence, State Key Laboratory of Virtual Reality Technology and Systems, Beihang University, Beijing 100191, China · School of Cyber Science and Technology, Shenzhen Campus of Sun Yat-Sen University, Shenzhen 518107, China
Do These Violent Delights Have Violent Ends? Measuring the Post-Merge Fate of Agentic Code
Jul 10, 2026Chunqiu Steven Xia, Courtney MillerPull Requests Coding Agents
University of Illinois Urbana-Champaign · Urbana, IL, USA · George Washington University +1
Co-occurring associated retained concepts in Diffusion Unlearning
Jun 23, 2026Miso Kim, Georu Lee, Yunji Kim +3Concept Erasure Child Sexual Abuse Material
Dongguk University-Seoul · Chung-Ang University · Korea Institute for Advanced Study
Mobile Pedipulation for Object Sliding via Hierarchical Control on a Wheeled Bipedal Robot
Jun 17, 2026Yue Qin, Yulun Zhuang, Zelin Shen +1Legged Robots Slipsense
Department of Robotics, University of Michigan, Ann Arbor, MI 48109, USA
The Market in the Model: Latent Diffusion as Neural Economy
Jun 17, 2026Eryk SalvaggioLatent Diffusion Model Child Sexual Abuse Material
Cambridge Digital Humanities, University of Cambridge · Machine Visual Culture Research Group, Max Planck Institute
Classifying by Proxy: Explainable and Reproducible Ensemble of Proxy Tasks for Child Sexual Abuse Imagery Classification
Jun 14, 2026Clara Ernesto, Carlos Caetano, Sandra Avila +3Child Sexual Abuse Material Reproducibility
Instituto de Ciências Matemáticas e de Computação (ICMC), Universidade Estadual de São Paulo (USP), Brazil · Instituto de Computação (IC), Universidade Estadual de Campinas (UNICAMP), Brazil · Departamento de Ciência da Computação, Universidade Federal de Minas Gerais (UFMG), Brazil +1
Position: Preventing AI-Generated CSAM Necessitates New Approaches to AI Safety
Jun 9, 2026Neil Kale, Rebecca Portnoff, Pratiksha Thaker +5Artificial Intelligence Safety Child Sexual Abuse Material
1Carnegie Mellon University · 2Thorn
ViLegalNLI: Natural Language Inference for Vietnamese Legal Texts
Apr 30, 2026Nhung Thi-Hong Duong, Mai Ngoc Ho, Tin Van Huynh +1Legal Reasoning Tasks Natural Language Inference
University of Information Technology, Ho Chi Minh City, Vietnam. · Vietnam National University, Ho Chi Minh City, Vietnam.
Evaluation without Generation: Non-Generative Assessment of Harmful Model Specialization with Applications to CSAM
Apr 28, 2026Vinith M. Suriyakumar, Ayush Sekhari, Lena Stempfle +5Adversarial Evaluation Child Sexual Abuse Material
MIT · BU · Thorn

Sep 29, 2026, cs.CVJ/K move · Enter open · S save

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

Arian Komaei Koma, Seyed Amir Kasaei, Aida Aryafar, Matin Ghiasi+5

Sharif University of Technology · Hong Kong University of Science and Technology

Text-To-Image Diffusion Models Diffusion Models Text-To-Image Exact Unlearning Concept Erasure Child Sexual Abuse Material+2

Open paper AskPDF arXiv

Machine unlearning has emerged as a critical post-hoc safety measure to erase sensitive concepts from Text-to-Image (T2I) models without prohibitive retraining. However, we reveal that current state-of-the-art (SOTA) approaches are brittle due to a severe lack of robustness to noise initialization. We call this phenomenon ``probabilistic forgetting'': suppressed concepts re-emerge under specific random initial noise conditions, despite appearing unlearned on other initializations. We trace this failure to the misalignment between standard Gaussian sampling during unlearning and the unlearning objective. Since the target concept manifests only in specific initial noise regions throughout the unlearning phase, uniform random sampling yields sparse, uninformative gradient updates that fail to drive robust erasure. To overcome this issue, we propose an adaptive, concept-conditioned sampling strategy that dynamically concentrates gradient updates on regions where the target concept manifests, down-weighting uninformative areas. We integrate our framework with six distinct SOTA unlearning methods across four diffusion backbones and evaluate it across safety, object, and artistic-style unlearning, as well as under black-box and white-box adversarial attacks. Our method reduces the conditional nudity re-emergence rate across random initializations by 67.2% on average over four baselines and lowers attack success rates across both adversarial evaluations. Across concept domains, Adaptive Noise Sampling strengthens adversarial robustness and non-target retention while preserving competitive generative quality and target-erasure performance.

Momentum

Related topics

Latest papers 14