ByteDance-Seed/Multi-SWE-bench_trajs

Multi-SWE-bench轨迹数据集存储了在Multi-SWE-bench排行榜上评估的所有代理生成的轨迹和日志。这个数据集用于多语言问题解决基准测试，包含了多种语言环境中代理的交互数据，可用于研究和评估代理在处理软件缺陷报告时的性能。

This dataset, Multi-SWE-bench Trajectories, contains all the trajectories and logs generated by agents evaluated on the Multi-SWE-bench leaderboard. It is used for a multilingual benchmark for issue resolving, comprising interaction data of agents in various linguistic environments, which can be used to research and evaluate the performance of agents in handling software bug reports.