Detecting CSAM Text-to-Image LoRAs From Weights
arXiv:2607.25750v1 Announce Type: cross Abstract: Low-rank adaptation (LoRA) fine-tuning has made it cheap and easy to customize open-weight image generation models for specific tasks, including the production of child sexual abuse material (CSAM). Existing moderation relies on metadata or generated outputs, but metadata can be deceptive and generating outputs may itself be unacceptable or illegal. We show that a safer signal lives in the weights. The top-left singular vectors of a LoRA's update
Record details
Published: 29 July 2026
Source: arXiv cs.CY
Category: Research
Topics: Children & education
Retrieved: 29 July 2026
Related evidence
These records share source-supplied organisations, an exact publisher byline, automatic topics or regions. The reason is shown on every link; related does not mean supporting, agreeing with or verifying this record.
Generative AI Availability, Grades, and Student Satisfaction at a Large University
arXiv cs.CY · 29 July 2026
Less Deliberate in Teams: Student LLM Use Across Individual and Collaborative Work
arXiv cs.CY · 29 July 2026
Effort Matters in Score-Based Admissions: How Retaking and Aggregation Shape Test Scores
arXiv cs.CY · 29 July 2026
Empathy and the Human-Moment Gaps of AI Chatbots: Insights from Empathy Displacement Theory
arXiv cs.CY · 29 July 2026
From Idea to Classroom in Days: Using "Vibe Coding" to Create a Programming Process Visualizer from IDE Activity Logs
arXiv cs.CY · 29 July 2026
Patterns of Learner-AI Interaction and Academic Performance in an Object-Oriented Programming Course
arXiv cs.CY · 29 July 2026
How to cite this record
ethics.ai (29 July 2026), “Detecting CSAM Text-to-Image LoRAs From Weights,” evidence record 14106, https://ethics.ai/record/14106 (originally published by arXiv cs.CY).
Use and limitations
This page is a stable index and citation surface for a source record. ethics.ai did not author the underlying report and has not independently verified every claim. Automatic topics may be imperfect. For consequential use, quote and cite the original publisher.