RedBench: A Universal Dataset for Comprehensive Red Teaming of Large Language Models

发布

2026年04月20日

采集 2026年04月20日 06:45

学术前沿 7.0 分 — Comprehensive red teaming dataset aggregating 37 benchmarks

原文： cs.CL updates on arXiv.org

评分 7.0 · 来源：cs.CL updates on arXiv.org · 发布于 2026-04-20

评分依据：Comprehensive red teaming dataset aggregating 37 benchmarks

GTA-2: Benchmarking General Tool Agents from Atomic Tool-Use to Open-Ended Workflows

Do LLMs Really Know What They Don't Know? Internal States Mainly Reflect Knowledge Recall Rather Than Truthfulness