CLASHarness Benchmarks Multimodal Agents in Multi-Agent Conflicts
A research consortium involving Google and six universities released CLASHarness, an open benchmark for testing 15 models when teammate reports conflict with system evidence. According to New Zhiyuan, its 600 executable tasks use high-resolution visual evidence and cover 20 domains; the report says no model exceeded a 30% end-to-end pass rate, with active verification and task handoff remaining major weaknesses. The generation pipeline, evaluation tasks, and an interactive trajectory viewer are available as open-source materials.
CLASHarness addresses a gap in existing computer-use benchmarks by testing whether multimodal agents can verify conflicting teammate information, repair system state, and complete collaboration handoffs.
新智元 2026-10-31 20:32 北京
新智元报道
现在的AI Agent已经越来越多地被组队部署到写代码、云运维、财务对账和企业审批等真实工作流中。
但现有的Computer-Use基准（如OSWorld、WebArena）几乎都默认了一个理想前提：环境信息永远一致，所有队友Agent都百分之百靠谱。
真实场景显然不是这样。在多Agent协作中，子Agent经常会因为读了过期缓存而汇报错误状态，或者在文件锁和权限上跟主Agent撞车。
#CLASHarness#Google#Emory University#University of Southern California#University of Florida#University of Chicago#University of California, Santa Barbara#Columbia University