Healer uses LLMs to dynamically generate and execute runtime error-handling code, with GPT-4 recovering from 72.8% of errors across four datasets.
In ChatGPT We Trust? Measuring and Characterizing the Reliability of ChatGPT.CoRR abs/2304.08979
6 Pith papers cite this work. Polarity classification is still indexing.
6
Pith papers citing it
verdicts
UNVERDICTED 6representative citing papers
PopQuiz Attack infers LLM training data membership by turning examples into quiz questions and measuring answer accuracy, reaching 0.873 average ROC-AUC across six models and outperforming prior methods by 20.6%.
Real-world jailbreak prompts collected from the wild achieve up to 0.95 attack success rates against major LLMs including GPT-4, with some persisting for over 240 days.
People show little ability to distinguish AI-generated from human-written health advice in online communities, with detection varying by health condition and unreliable heuristics.
GUARD automates generation of guideline-violating questions and jailbreak diagnostics to test LLM compliance with government ethics guidelines, validated empirically on eight models and extended to vision-language models.
Analysis of ClawHub shows language-based functional divides in agent skills, with over 30% flagged suspicious and submission-time documentation enabling 73% accurate risk prediction.
citing papers explorer
-
Towards Agentic Runtime Healing
Healer uses LLMs to dynamically generate and execute runtime error-handling code, with GPT-4 recovering from 72.8% of errors across four datasets.
-
Pop Quiz Attack: Black-box Membership Inference Attacks Against Large Language Models
PopQuiz Attack infers LLM training data membership by turning examples into quiz questions and measuring answer accuracy, reaching 0.873 average ROC-AUC across six models and outperforming prior methods by 20.6%.
-
"Do Anything Now": Characterizing and Evaluating In-The-Wild Jailbreak Prompts on Large Language Models
Real-world jailbreak prompts collected from the wild achieve up to 0.95 attack success rates against major LLMs including GPT-4, with some persisting for over 240 days.
-
Discerning Authorship in Online Health Communities: Experience, Trust, and Transparency Implications for Moderating AI
People show little ability to distinguish AI-generated from human-written health advice in online communities, with detection varying by health condition and unreliable heuristics.
-
GUARD: Guideline Upholding Test through Adaptive Role-play and Jailbreak Diagnostics for LLMs
GUARD automates generation of guideline-violating questions and jailbreak diagnostics to test LLM compliance with government ethics guidelines, validated empirically on eight models and extended to vision-language models.
-
Red Skills or Blue Skills? A Dive Into Skills Published on ClawHub
Analysis of ClawHub shows language-based functional divides in agent skills, with over 30% flagged suspicious and submission-time documentation enabling 73% accurate risk prediction.