Pinned Loading

1. aisa-group/PostTrainBench aisa-group/PostTrainBench Public

   Measuring how well CLI agents like Claude Code or Codex CLI can post-train base LLMs on a single H100 GPU in 10 hours

   Python 537 61

2. tml-epfl/os-harm tml-epfl/os-harm Public

   OS-Harm: A Benchmark for Measuring Safety of Computer Use Agents [NeurIPS 2025 Spotlight]

   Jupyter Notebook 71 6

3. tml-epfl/llm-past-tense tml-epfl/llm-past-tense Public

   Does Refusal Training in LLMs Generalize to the Past Tense? [ICLR 2025]

   Python 78 12

4. tml-epfl/llm-adaptive-attacks tml-epfl/llm-adaptive-attacks Public

   Jailbreaking Leading Safety-Aligned LLMs with Simple Adaptive Attacks [ICLR 2025]

   Shell 393 47

5. JailbreakBench/jailbreakbench JailbreakBench/jailbreakbench Public

   JailbreakBench: An Open Robustness Benchmark for Jailbreaking Language Models [NeurIPS 2024 Datasets and Benchmarks Track]

   Python 660 78

6. RobustBench/robustbench RobustBench/robustbench Public

   RobustBench: a standardized adversarial robustness benchmark [NeurIPS 2021 Benchmarks and Datasets Track]

   Python 782 106