TianchenGuan/ComponentBench download history
TianchenGuan/ComponentBench is an other dataset on the Hugging Face Hub. In the last 30 days it was downloaded 418 times (165 in the last 7 days), and 1,717 times in total. It ranks #38,105 among datasets by monthly downloads.
ComponentBench Diagnosing Component-Level Failures in Computer-Use Agents ComponentBench is a diagnostic benchmark for computer-use agents that targets the middle layer between atomic GUI-grounding tests (e.g., ScreenSpot) and long-horizon workflow benchmarks (e.g., WebArena, OSWorld)