Concept-Level Risk and Calibration for Governance in Diffusion Foundation Models
Open Access
- 9 November 2026
- conference paper
- Published by Association for Computing Machinery (ACM)
- p. 12565-12574
Abstract
No abstract availableFunding Information
- National Natural Science Foundation of China (62522112)
- Outstanding Youth Fund Program of Jiangxi Province (20252BAC220008)
- Jiangxi Key Research and Development Program (20261BCE310050)
- Ganpo Talent Program of Jiangxi Province (gpyc20240012)
- China Scholarship Council Program (202506830129)
This publication has 33 references indexed in Scilit:
- ConceptGuard: Continual Personalized Text-to-Image Generation with Forgetting and Confusion Mitigation2025
- Memories of Forgotten Concepts2025
- T2ISafety: Benchmark for Assessing Fairness, Toxicity, and Privacy in Image Generation2025
- Six-CD: Benchmarking Concept Removals for Text-to-image Diffusion Models2025
- Mind the Trojan Horse: Image Prompt Adapter Enabling Scalable and Deceptive Jailbreaking2025
- MACE: Mass Concept Erasure in Diffusion Models2024
- Null-text Inversion for Editing Real Images using Guided Diffusion Models2023
- DreamBooth: Fine Tuning Text-to-Image Diffusion Models for Subject-Driven Generation2023
- Safe Latent Diffusion: Mitigating Inappropriate Degeneration in Diffusion Models2023
- High-Resolution Image Synthesis with Latent Diffusion Models2022