• Adaptive Stress Testing for Language Model Toxicity

  • Jan 20 2025
  • Durée: 15 min
  • Podcast

Adaptive Stress Testing for Language Model Toxicity

Écouter gratuitement

Voir les détails du balado

  • Résumé

  • This episode explores ASTPrompter, a novel approach to automated red-teaming for large language models (LLMs). Unlike traditional methods that focus on simply triggering toxic outputs, ASTPrompter is designed to discover likely toxic prompts – those that could naturally emerge during regular language model use. The approach uses Adaptive Stress Testing (AST), a technique that identifies likely failure points, and reinforcement learning to train an "adversary" model. This adversary generates prompts that aim to elicit toxic responses from a "defender" model, but importantly, these prompts have a low perplexity, meaning they are realistic and likely to occur, unlike many prompts generated by other methods.
    Voir plus Voir moins
Voir plus Voir moins

Ce que les auditeurs disent de Adaptive Stress Testing for Language Model Toxicity

Moyenne des évaluations de clients

Évaluations – Cliquez sur les onglets pour changer la source des évaluations.

Évaluations sur Audible.ca

Évaluations sur Amazon.ca
Il n'y a pas encore de critiques pour ce titre.