Aggregator
NDSS 2025 – Black-Box Membership Inference Attacks Against Fine-Tuned Diffusion Models
Session 12C: Membership Inference
Authors, Creators & Presenters: Yan Pang (University of Virginia), Tianhao Wang (University of Virginia)
PAPER
Black-box Membership Inference Attacks against Fine-tuned Diffusion Models
With the rapid advancement of diffusion-based image-generative models, the quality of generated images has become increasingly photorealistic. Moreover, with the release of high-quality pre-trained image-generative models, a growing number of users are downloading these pre-trained models to fine-tune them with downstream datasets for various image-generation tasks. However, employing such powerful pre-trained models in downstream tasks presents significant privacy leakage risks. In this paper, we propose the first scores-based membership inference attack framework tailored for recent diffusion models, and in the more stringent black-box access setting. Considering four distinct attack scenarios and three types of attacks, this framework is capable of targeting any popular conditional generator model, achieving high precision, evidenced by an impressive AUC of 0.95.
ABOUT NDSS
The Network and Distributed System Security Symposium (NDSS) fosters information exchange among researchers and practitioners of network and distributed system security. The target audience includes those interested in practical aspects of network and distributed system security, with a focus on actual system design and implementation. A major goal is to encourage and enable the Internet community to apply, deploy, and advance the state of available security technologies.
Our thanks to the Network and Distributed System Security (NDSS) Symposium for publishing their Creators, Authors and Presenter’s superb NDSS Symposium 2025 Conference content on the Organizations' YouTube Channel.
The post NDSS 2025 – Black-Box Membership Inference Attacks Against Fine-Tuned Diffusion Models appeared first on Security Boulevard.
SecWiki News 2026-02-14 Review
评估与增强大语言模型解决夺旗赛挑战的能力 by ourren
PentestAgent:将LLM代理融入自动化渗透测试 by ourren
2025年全球身份认证攻击风险研究报告 by ourren
JoySafeter:AI驱动安全(AISecOps)新范式 by ourren
更多最新文章,请访问SecWiki
U.S. CISA adds a flaw in BeyondTrust RS and PRA to its Known Exploited Vulnerabilities catalog
Такси помнит, где вы изменяли: как навигаторы и каршеринг сливают ваши маршруты... и как себя обезопасить
小心AI推荐投毒 (AI Recommendation Poisoning)
Snail mail letters target Trezor and Ledger users in crypto-theft attacks
Математики 100 лет не могли описать текущую лаву. Но теперь все получилось
Waymo 付费给 DoorDash 零工给自动驾驶汽车关车门
OpenAI 再次表示 DeepSeek 利用蒸馏训练其模型
Треснувшая скорлупа Вселенной. Хаббл снял, как умирающая звезда вспарывает тьму лучами света
Ars 报道的 AI 新闻包含了 AI 生成的虚假引言
得克萨斯大学圣安东尼奥分校 | 通过代码生成LLM对软件包幻觉进行全面分析
CVE-2026-1096 | Best-wp-google-map Plugin up to 2.1 on WordPress Shortcode latitude/longitudinal cross site scripting
CVE-2026-0751 | Payment Page Form for Stripe Plugin up to 1.4.6 on WordPress pricing_plan_select_text_font_family cross site scripting
Seedance 2.0 炸场之后,豆包 Seed2.0 能否再度勇攀高峰?
G.O.S.S.I.P 2026 新春总动员(1):512字节实现一个C语言编译器!
Парализованные пациенты внезапно идут на поправку — виновны клетки, которые все считали бесполезными
The Relationship Tax: What Staying With Your SEG Is Really Costing Your MSP
Part 3 of a 4-Part Series: The SEG Breakup Guide for MSPs
The post The Relationship Tax: What Staying With Your SEG Is Really Costing Your MSP appeared first on Security Boulevard.