AI 互相出题考对方，基准测试从此不会过时
代码基准测试有个死穴：题目一发布，迟早被 AI 背下来，测不出真本事。这篇干脆让 AI 互相出题：红方在真实代码仓库里写新功能、故意埋一个隐蔽 bug，蓝方不知情地来修，修对了才算赢，然后互换角色。11 个模型打了 1500 轮，最强解题手 GPT-5.5 能修好 98.3% 的 bug，最强出题手 Claude Opus 4.8 的题让对手翻车率 27.3%。更关键的是，新模型出的题比旧模型难得多——42.9% 对 4.0% 的击败率，说明题目难度会跟着出题者能力自动水涨船高。这意味着以后不用再担心测试集被刷爆，AI 的进步速度会由 AI 自己来丈量。
📄 原文摘要(英文)
Static coding benchmarks are fixed at release. Over time their tasks leak into training data, and stronger agents eventually saturate them. We argue that a benchmark should instead aim to be self-scaling: its tasks and oracles are created at evaluation time, so no content exists in advance to leak, and its difficulty is set by the participating agents rather than fixed at release. We introduce SWE-Duel, an arena in which two coding agents evaluate each other over mirrored Red-Blue turns. In each turn, the Red agent works on a real repository across three context-isolated sessions. It proposes and implements a new feature with feature tests, then embeds a subtle bug in that feature along with sealed bug-triggering tests. The resulting pull request is admitted only if (1) it passes the feature tests and the repository's existing tests while failing exactly the bug tests, and (2) a fresh session of the same agent can repair the bug blind, which rules out unsolvable tasks. The Blue agent then reviews the pull request without being told a bug exists. It wins only if its revision passes all three test suites, and then the two agents swap roles. We run a tournament among 11 model-harness participants on 15 repositories in five languages for 1,500 turns. The leaders win in different ways. GPT-5.5 with Codex CLI ranks first as the strongest solver, repairing 98.3% of the defects it reviews. Claude Opus 4.8 with mini-swe-agent ranks second as the strongest task author, with a 27.3% Red win rate. Finally, within one model family, tasks written by the newest version defeat a fixed-reference solver far more often than those of older versions (42.9% vs. 4.0%), giving preliminary evidence that task difficulty scales with task authoring capability.