Trang chủ / Tin tức / Khoa học & công nghệ
Khoa học & công nghệ

Concept-Level Risk and Calibration for Governance in Diffusion Foundation Models

05/11/2026 · Thành viên: PGS.TS Nguyễn Chấn Hùng · Nguồn: OpenAlex

Diffusion models have become a core paradigm for multimedia generation, offering powerful concept-driven controllability for personalization, semantic editing, and selective unlearning. However, as...

<p><strong>Tác giả:</strong> Kun Xu, Yushu Zhang, Tao Wang, Shuren Qi, Barbara Carminati, Elena Ferrari, Yuming Fang</p>

<p>Diffusion models have become a core paradigm for multimedia generation, offering powerful concept-driven controllability for personalization, semantic editing, and selective unlearning. However, as semantic control extends beyond natural-language prompts to learned embeddings and intervention pipelines, the safety and governance of these systems become increasingly difficult to evaluate in a unified manner, especially for safety-sensitive, identity-linked, and other privacy-relevant concepts....</p>

<p>📄 DOI: <a href="https://doi.org/https://doi.org/10.1145/3767308.3835863" target="_blank" rel="noopener">https://doi.org/10.1145/3767308.3835863</a></p>

<p>📖 <a href="https://doi.org/https://doi.org/10.1145/3767308.3835863" target="_blank" rel="noopener">Xem bài báo đầy đủ</a></p>

Bài viết được biên soạn lại bởi LIFVietnam. Nguồn tham khảo: OpenAlex