[HTML][HTML] A survey on large language model (llm) security and privacy: The good, the bad, and the ugly
Abstract Large Language Models (LLMs), such as ChatGPT and Bard, have revolutionized
natural language understanding and generation. They possess deep language …
natural language understanding and generation. They possess deep language …
An archival perspective on pretraining data
Alongside an explosion in research and development related to large language models,
there has been a concomitant rise in the creation of pretraining datasets—massive …
there has been a concomitant rise in the creation of pretraining datasets—massive …
Stealing part of a production language model
We introduce the first model-stealing attack that extracts precise, nontrivial information from
black-box production language models like OpenAI's ChatGPT or Google's PaLM-2 …
black-box production language models like OpenAI's ChatGPT or Google's PaLM-2 …
Sok: Memorization in general-purpose large language models
Large Language Models (LLMs) are advancing at a remarkable pace, with myriad
applications under development. Unlike most earlier machine learning models, they are no …
applications under development. Unlike most earlier machine learning models, they are no …
Privacy backdoors: Enhancing membership inference through poisoning pre-trained models
It is commonplace to produce application-specific models by fine-tuning large pre-trained
models using a small bespoke dataset. The widespread availability of foundation model …
models using a small bespoke dataset. The widespread availability of foundation model …
User inference attacks on large language models
Fine-tuning is a common and effective method for tailoring large language models (LLMs) to
specialized tasks and applications. In this paper, we study the privacy implications of fine …
specialized tasks and applications. In this paper, we study the privacy implications of fine …
A False Sense of Safety: Unsafe Information Leakage in'Safe'AI Responses
Large Language Models (LLMs) are vulnerable to jailbreaks $\unicode {x2013} $ methods
to elicit harmful or generally impermissible outputs. Safety measures are developed and …
to elicit harmful or generally impermissible outputs. Safety measures are developed and …
Cognitive Tracing Data Trails: Auditing Data Provenance in Discriminative Language Models Using Accumulated Discrepancy Score
The burgeoning practice of unauthorized acquisition and utilization of personal textual data
(eg, social media comments and search histories) by certain entities has become a …
(eg, social media comments and search histories) by certain entities has become a …
REVS: Unlearning Sensitive Information in Language Models via Rank Editing in the Vocabulary Space
T Ashuach, M Tutek, Y Belinkov - arXiv preprint arXiv:2406.09325, 2024 - arxiv.org
Large language models (LLMs) risk inadvertently memorizing and divulging sensitive or
personally identifiable information (PII) seen in training data, causing privacy concerns …
personally identifiable information (PII) seen in training data, causing privacy concerns …
Large language models in 6G security: challenges and opportunities
The rapid integration of Generative AI (GenAI) and Large Language Models (LLMs) in
sectors such as education and healthcare have marked a significant advancement in …
sectors such as education and healthcare have marked a significant advancement in …