FrontierChallenge: Evaluating Scientific Workflow Completion
Scientific agents increasingly analyze data, execute code, and produce research artifacts, yet most benchmarks emphasize final answers, isolated programs, or a single domain. We introduce FrontierChallenge, a cross-domain benchmark comprising 300 end-to-end scientific workflows. In this paper, we release and evaluate 9...
Liangcai Su, Zhao-Peng Feng, Zhuo Chen et al.
· 0 citations