Apple 论文：工具使用智能体的泛化更依赖应用与数据库多样性，而非任务数量
Apple 论文发现工具使用智能体的泛化更依赖应用与数据库多样性而非任务数量
Apple 论文发现，随训练数据增长，工具使用智能体从多样化应用和数据库状态中获得的泛化收益大于增加任务数量。团队构建 SynthAPI 流水线，用 LLM 编写带真实 Python 工具和数据库的合成应用并自动校验任务成败，跨 849 个训练模型改变应用、任务和数据库版本。
Rohan Paul· @rohanpaul_ai · X·2026-10-29 12:11· 33 分钟前AI 评分52
AI 导读
Apple 论文研究合成训练数据配比对 LLM 工具使用智能体泛化的影响，发现随数据预算增长，扩展应用和数据库多样性比增加任务数带来更大收益。团队构建 SynthAPI 流水线，用 LLM 编写带真实 Python 工具和数据库的合成应用并自动检查任务成败，跨 849 个训练模型改变应用、任务和数据库版本。
正文 · 原文
New Apple paper finds that, as training data grows, tool-use agents gain more from diverse apps and database states than from more tasks.
that the mix of synthetic training data, not just its size, decides how well tool-use agents generalize to new benchmarks.
Teams train tool-use agents on synthetic environments, and most pipelines focus on generating more tasks. It was unclear whether new tools, new tasks, or new starting data matter most.
They built SynthAPI, which has an LLM code synthetic apps with real Python tools and databases, then write tasks with automatic success checks. They varied apps, tasks, and database versions across 849 trained models.
At a fixed 32,768-task budget, more apps and database versions lifted a 4B model on AppWorld's challenge split from 38.4% to 45.0%. Their scaling law, which tracks data mix, predicted larger held-out runs with 0.0285 error, versus 0.1706 for a task-count-only law.
来源：Rohan Paul · x.com
来源：RadarAI · aihot.news