Meta 与卡内基梅隆大学论文：RL 训练的小模型选工具准确率超过 402B Llama-4-Maverick
New Meta and Carnegie Mellon paper finds that small models trained with RL pick the right tools more often than a prompted 402B Llama-4-Maverick.
Meta 与卡内基梅隆大学的新论文发现，用强化学习训练的小模型在工具选择上比提示词驱动的 402B Llama-4-Maverick 更常选对工具。研究者训练 Qwen3 模型把请求拆成步骤、逐步用关键词搜索并每步选 1 个工具，训练先用教师编写的示例，再用强化学习奖励有效的搜索步骤和精确选择。在 ToolBench 上，训练后的 8B 模型精确选对工具的比例为 41.5%，而 Maverick 最佳提示词设置为 30.6%；未经重新训练，它在 4 个新工具基准中的 3 个上超过 Maverick，并在换用更强的检索器后进一步提升。
推荐理由
AI 评分 60
原文给出了小模型与 402B 大模型在工具选择上的具体对比数字，可供评估智能体工具检索方案时参考。
主题
正文
New Meta and Carnegie Mellon paper finds that small models trained with RL pick the right tools more often than a prompted 402B Llama-4-Maverick.
Agents can face more than 10,000 tools, and their descriptions cannot all fit in context. Retrievers only rank by relevance, so recent setups have Claude or ChatGPT pick the tools.
They trained Qwen3 models to split a request into steps, search each step with plain keyword search, and pick 1 tool per step. Training used teacher-written examples, then RL that rewards useful search steps and exact picks.
On ToolBench, the trained 8B model picked exactly the right tools 41.5% of the time, versus 30.6% for Maverick's best prompted setup. With no retraining, it beat Maverick on 3 of 4 new tool benchmarks and improved when given a stronger retriever.
If your agent chooses from a large tool catalog, an RL-trained small model can handle selection instead of a prompted large one.