Reproduce “OR-License: Evaluating LLM Agents as Real-World Operations Research Practitioners” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Peisong Wang, Feixue Liu, Yize Meng, Bowen Liu, Ruoyu Chen, Gaoyang Dai, Xing He, Aiqian Li and 11 more
OR-License evaluates agents on industrial operations-research tasks. It gives agents de-identified engagement materials and lets them ask a simulated client for business facts, while leaving modeling and solving to the agents. On 31 tasks, Grok 4.6 (high) achieves 86.55% DAQ, below the expert reference of 100%; its strict feasibility is 88.39%, within this benchmark and reported harness.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.