Threats, attacks, and defenses in machine unlearning: A survey
Machine Unlearning (MU) has recently gained considerable attention due to its potential to
achieve Safe AI by removing the influence of specific data from trained Machine Learning …
achieve Safe AI by removing the influence of specific data from trained Machine Learning …
Improving alignment and robustness with circuit breakers
AI systems can take harmful actions and are highly vulnerable to adversarial attacks. We
present an approach, inspired by recent advances in representation engineering, that …
present an approach, inspired by recent advances in representation engineering, that …
An adversarial perspective on machine unlearning for ai safety
Large language models are finetuned to refuse questions about hazardous knowledge, but
these protections can often be bypassed. Unlearning methods aim at completely removing …
these protections can often be bypassed. Unlearning methods aim at completely removing …
Trustworthy, responsible, and safe ai: A comprehensive architectural framework for ai safety with challenges and mitigations
AI Safety is an emerging area of critical importance to the safe adoption and deployment of
AI systems. With the rapid proliferation of AI and especially with the recent advancement of …
AI systems. With the rapid proliferation of AI and especially with the recent advancement of …
Efficient adversarial training in llms with continuous attacks
Large language models (LLMs) are vulnerable to adversarial attacks that can bypass their
safety guardrails. In many domains, adversarial training has proven to be one of the most …
safety guardrails. In many domains, adversarial training has proven to be one of the most …
Soft prompts go hard: Steering visual language models with hidden meta-instructions
We introduce a new type of indirect injection attacks against language models that operate
on images: hidden''meta-instructions''that influence how the model interprets the image and …
on images: hidden''meta-instructions''that influence how the model interprets the image and …
Multilingual blending: Llm safety alignment evaluation with language mixture
As safety remains a crucial concern throughout the development lifecycle of Large
Language Models (LLMs), researchers and industrial practitioners have increasingly …
Language Models (LLMs), researchers and industrial practitioners have increasingly …
Uncovering safety risks of large language models through concept activation vector
Despite careful safety alignment, current large language models (LLMs) remain vulnerable
to various attacks. To further unveil the safety risks of LLMs, we introduce a Safety Concept …
to various attacks. To further unveil the safety risks of LLMs, we introduce a Safety Concept …
A probabilistic perspective on unlearning and alignment for large language models
Comprehensive evaluation of Large Language Models (LLMs) is an open research problem.
Existing evaluations rely on deterministic point estimates generated via greedy decoding …
Existing evaluations rely on deterministic point estimates generated via greedy decoding …
Evaluating the cybersecurity robustness of commercial llms against adversarial prompts: A promptbench analysis
T Goto, K Ono, A Morita - Authorea Preprints, 2024 - techrxiv.org
This study presents a comprehensive evaluation of the cybersecurity robustness of five
leading Large Language Models (LLMs)-ChatGPT-4, Google Gemini, Anthropic Claude …
leading Large Language Models (LLMs)-ChatGPT-4, Google Gemini, Anthropic Claude …