[1]
Ni, Z. et al. 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 (Mar. 2026), 32564–32572. DOI:https://doi.org/10.1609/aaai.v40i38.40533.