Meta/CMU Zanette 模型：小模型也能选工具且可迁移，8B 41.5% > Maverick 30.6%

View PDF HTML (experimental)

Abstract:As LLM agents operate over corpora of more than ten thousand tools, selecting the right tool for a complex multi-step user instruction becomes a central bottleneck. Tool descriptions cannot all fit in context, and relevance-based retrievers only rank tools without knowing which are actually needed. Recent work instead leverages proprietary commercial LLMs like Claude or ChatGPT to drive a retriever, decomposing the instruction and selecting over retrieved candidates. We show that a small, open model can be trained via reinforcement learning to perform the same functionality. On ToolBench, a Qwen3-8B model post-trained with our recipe outperforms Llama-4-Maverick (402B). In addition, the trained model transfers directly to different retrievers and generalizes to new tool corpora in different domains.

Submission history

From: Zhaoyi Zhou [view email]
[v1] Thu, 22 Oct 2026 16:07:49 UTC (190 KB)