Model stealing, distillation attacks, and weight extraction
Newest first · 3 reviewed on this page
Search insteadJunxi Chen, Junhao Dong, Xiaohua Xie
Recent work has demonstrated the potential of contrastive steering for jailbreaking Large Language Models (LLMs). However, existing methods rely on limited and inherently biased contrastive prompts and require laborious manual tuning of steering strength, limiting their…
Shuze Liu, Qianwen Guo, Yushun Dong
Large language models (LLMs) are increasingly deployed through hosted APIs, making model extraction a practical threat to model ownership and service security. However, individual extraction queries often resemble benign requests, and existing evaluations often focus on…
Yuyang Dai, Yushun Dong
Large language models deployed as commercial APIs are vulnerable to model extraction attacks, while existing defenses either act too late or degrade utility for legitimate users. We propose \textbf{Knowledge Trap}, a defense that redirects extraction attacks toward…
Chen Chen, Yao-Lin Chen, Xue-Han Sun +5
Large language model (LLM) judges are increasingly used across various evaluation scenarios, making their judgment capabilities valuable intellectual property. However, black-box access exposes these capabilities to model extraction attacks. Existing extraction methods do not…
F. Schüssler, S. Bisero, M. Cellier +6
We present a new Astro-COLIBRI component that converts free-text GCN Circulars into structured, event-linked follow-up records and combines them with structured reports submitted directly by the community. A continuously running Circular listener associates new reports with…
Alessandra Mancas, Mounir Ammam, Hyacinth Ali +2
Large language models (LLMs) have recently shown strong capabilities for code understanding, making them promising for reverse engineering domain models from source code. However, state-ofthe- art proprietary LLMs cannot be used in many industrial contexts due to privacy and…
Ruinan Ma, Yu-Xi Ma, Meng-Xia Ren +3
Rizwan Tanveer
Background. Adversarial machine learning has progressed from a marginal concern within machine learning research into a first-order discipline for the secure deployment of artificial intelligence systems in regulated and operational environments. The contemporary threat…
Zhiyu Liao, Kang Chen, Y. Lin +5
Large Language Models (LLMs) have become central to numerous natural language processing tasks, but their vulnerabilities present significant security and ethical challenges. This systematic survey explores the evolving landscape of attack and defense techniques in LLMs. We…
Xinyue Shen, Yiting Qu, Michael Backes +1
Demonstrates attacks that steal the prompts used to generate images from text-to-image models, raising IP and privacy concerns.
Nicholas Carlini, Daniel Paleka, Krishnamurthy Dj Dvijotham +10
Demonstrates that it is possible to steal the embedding projection layer of production LLMs like OpenAI's models through the API, confirming model extraction risks.
Milad Nasr, Nicholas Carlini, Jonathan Hayase +7
Develops a scalable attack to extract over a gigabyte of training data from semi-open and closed models including ChatGPT, at a cost of roughly $200.