Semantic manipulations of SKILL.md descriptions enable effective supply-chain attacks that bias AI agent skill registries toward adversarial skills in discovery, selection, and governance.
arXiv preprint arXiv:2402.15570 , year=
6 Pith papers cite this work. Polarity classification is still indexing.
6
Pith papers citing it
citation-role summary
background 1
method 1
citation-polarity summary
verdicts
UNVERDICTED 6representative citing papers
OTTER optimizes prompts to decouple surface toxicity from adversarial intent, raising attack success rates on GPT models from 7% to 84% across 457 AdvBench examples.
A safeguard that uses speculative inference on small language models to produce draft responses for safety prediction, lowering false negatives in pre-model jailbreak detection.
REALISTA generates semantically coherent adversarial prompts via latent-space optimization over input-dependent editing directions, achieving stronger hallucination elicitation than prior realistic attacks on open-source and reasoning LLMs.
Integrating pretrained sparse autoencoders into LLM residual streams reduces jailbreak success rates by up to 5x across multiple models and attacks.
LLM safety evaluations are hindered by noise in dataset curation, automated red-teaming, response generation, and LLM-judge evaluation, making fair comparisons difficult and slowing progress.
citing papers explorer
-
Under the Hood of SKILL.md: Semantic Supply-chain Attacks on AI Agent Skill Registry
Semantic manipulations of SKILL.md descriptions enable effective supply-chain attacks that bias AI agent skill registries toward adversarial skills in discovery, selection, and governance.
-
OTTER: A Red-Teaming System for Toxicity-Evading Jailbreak Prompt Optimization
OTTER optimizes prompts to decouple surface toxicity from adversarial intent, raising attack success rates on GPT models from 7% to 84% across 457 AdvBench examples.
-
Exploring and Developing a Pre-Model Safeguard with Draft Models
A safeguard that uses speculative inference on small language models to produce draft responses for safety prediction, lowering false negatives in pre-model jailbreak detection.
-
REALISTA: Realistic Latent Adversarial Attacks that Elicit LLM Hallucinations
REALISTA generates semantically coherent adversarial prompts via latent-space optimization over input-dependent editing directions, achieving stronger hallucination elicitation than prior realistic attacks on open-source and reasoning LLMs.
-
Towards Understanding the Robustness of Sparse Autoencoders
Integrating pretrained sparse autoencoders into LLM residual streams reduces jailbreak success rates by up to 5x across multiple models and attacks.
-
LLM-Safety Evaluations Lack Robustness
LLM safety evaluations are hindered by noise in dataset curation, automated red-teaming, response generation, and LLM-judge evaluation, making fair comparisons difficult and slowing progress.