From a176875dacd1fd9cfbf3318bebe99f0ac2cb6f1e Mon Sep 17 00:00:00 2001 From: Yuxuan Zhang Date: Wed, 20 May 2026 18:16:02 -0700 Subject: [PATCH] Add ClawBench (arXiv:2604.08523) to Benchmark --- README.md | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/README.md b/README.md index 70b8947..7bee993 100644 --- a/README.md +++ b/README.md @@ -319,6 +319,13 @@ A curated list of resources about AI agents for Computer Use, including research ### Benchmark +- [ClawBench: Can AI Agents Complete Everyday Online Tasks?](https://arxiv.org/abs/2604.08523) (May. 2026) + - Live-website browser-agent benchmark; 283 tasks (V1 153 + V2 130) across 163 live platforms (15 life categories) + - Two-stage scoring: HTTP-request interception at per-task URL/method schema (Stage 1) + LLM judge on intercepted payload (Stage 2) + - [Website](https://claw-bench.com) + - [Code](https://github.com/reacher-z/ClawBench) + - [Dataset](https://huggingface.co/datasets/NAIL-Group/ClawBench) + - [A3: Android Agent Arena for Mobile GUI Agents](https://arxiv.org/abs/2501.01149) (Jan. 2025) - Novel evaluation platform with 201 tasks across 21 widely used third-party apps - [Website](https://yuxiangchai.github.io/Android-Agent-Arena/)