AI 自我进化评测：企业任务上最高提升 16 个百分点，但离天花板还远
AI 智能体现在能「自我进化」:干完一个活儿，把经验存下来，下次遇到类似任务直接复用。听起来很美，但怎么证明它真学到了、而不是背了答案？这篇论文造了个专门测这个的考场，题目来自真实企业流程(CRM、ERP、财务、医疗、法律等)，把每个流程拆成原子规则，训练时给一部分，测试时重新组合成新任务——这样测试成绩的提升就只能归功于「进化」，而不是记忆。结果：四个主流智能体在自我进化后，准确率最高提升 16.44 个百分点，但最好的也远没到 91.6% 的「全知上限」。换句话说，AI 确实在进步，但离真正把经验用活还差得远。这不是你明天能用的功能，但它告诉你：别急着把重要业务交给 AI 自己「越用越聪明」，它现在还没那么聪明。
📄 原文摘要(英文)
Agent self-evolution updates an agent's persistent state from prior experience and reuses it to solve related tasks more effectively. Evaluating self-evolution is difficult: existing benchmarks provide limited coverage of economically valuable task domains, do not always design training and test tasks such that test-time gains can be attributed to training experience, and remain vulnerable to data contamination. We present GDPevo, an evolution-native benchmark grounded in GDP-related enterprise workflows, together with the fully automated data pipeline that generates it. Its core mechanism, rule hybridization, decomposes each enterprise workflow into atomic business rules, distributes subsets of these rules across training tasks, and recombines them in held-out test tasks so that test-time gains are attributable. GDPevo spans CRM, ERP, finance, healthcare, legal, and data-centric workflows. Its V1 release contains 120 tasks in 12 groups, with five training and five held-out test tasks per group. Full automation enables the pipeline to expand the suite to 240 tasks in 24 groups (V2) within two days, providing a practical response to contamination. Using GDPevo, we evaluate four agents, each comprising a harness and a model, under four supervision types. Self-evolution consistently improves held-out accuracy by up to 16.44 percentage points. But the best evolved agents remain far below the fully informed oracle ceiling of 91.6%, indicating that the self-evolution ability of current agents remains far from fully realized. We publicly release the pipeline, benchmark, and full evaluation results at https://github.com/Prism-Shadow/GDPevo.