MLCommons 发布 Jailbreak Benchmark v1.0
Get the story
2026年10月5日,MLCommons 在 arXiv 发布 Jailbreak Benchmark v1.0,提供评估大语言模型对单轮文本越狱攻击鲁棒性的端到端方法。该基准包含 264 个种子提示,覆盖 11 类风险,基于 MLCommons Jailbreak Taxonomy 评测了 8 个开放权重系统,并按 AILuminate Assessment Standard v1.4 进行评分。这是目前该事件的唯一报道,尚无后续更新或相互矛盾的信息。
Generated from reports · updated 16 hr ago
Timeline
Follow the coverage from different angles.
- arXiv · Artificial IntelligenceMLCommons Jailbreak Benchmark v1.0
MLCommons 发布 Jailbreak Benchmark v1.0,提供评估大语言模型对单轮文本越狱攻击鲁棒性的端到端方法。该基准用 264 个种子提示覆盖 11 类风险,基于 MLCommons Jailbreak Taxonomy 评测 8 个开放权重系统,并按 AILuminate Assessment Standard v1.4 评分。
Heat trend
Current heat 7·Comparable peak 10(Oct 5)·Comparable change over 24 hours –
The trend compares only the same participants observed continuously; its range may be smaller than the current heat count. Move or click on the chart to inspect hourly heat; use the left and right arrow keys to switch.