评分 7 · 来源：cs.AI updates on arXiv.org · 发布于 2026-04-14
评分依据：Fills gap in evaluating combined capabilities (coding + research + GUI) rather than isolated skills. Highly relevant as agents move toward unified architectures.
CocoaBench: Evaluating Unified Digital Agents in the Wild
发布
采集
学术前沿 7.0 分
— Fills gap in evaluating combined capabilities (coding + research + GUI) rather than isolated skills. Highly relevant as agents move toward unified architectures.