Ni, Z., Wang, H., Zhang, S., Lu, S., He, Z., , W., … Lyu, P. (2026). GitTaskBench: A Benchmark for Code Agents Solving Real-World Tasks Through Code Repository Leveraging. Proceedings of the AAAI Conference on Artificial Intelligence, 40(38), 32564–32572. https://doi.org/10.1609/aaai.v40i38.40533