Training Small Language Models for Transferable Tool Selection

Open the resource Read the PDF

The PDF opens on the site where the paper is published.

Authors: Zhaoyi Zhou, Akshayaa Magesh, Jeongyeol Kwon, Artem Kuriksha, Daniel Jiang, Andrea Zanette, Ayush Jain
Venue: arXiv (preprint, not yet peer-reviewed)
Published: 2026-10-22
ID: arXiv:2610.31483

How to cite it

Built from the details listed on this page. Check them against the paper before you submit.

APA

Zhou, Z., Magesh, A., Kwon, J., Kuriksha, A., Jiang, D., Zanette, A., & Jain, A. (2026). Training Small Language Models for Transferable Tool Selection. arXiv (preprint, not yet peer-reviewed). https://arxiv.org/abs/2610.31483

MLA

Zhou, Zhaoyi, et al. "Training Small Language Models for Transferable Tool Selection." arXiv (preprint, not yet peer-reviewed), 2026, https://arxiv.org/abs/2610.31483.

BibTeX

@article{zhou2026training,
  title = {Training Small Language Models for Transferable Tool Selection},
  author = {Zhaoyi Zhou and Akshayaa Magesh and Jeongyeol Kwon and Artem Kuriksha and Daniel Jiang and Andrea Zanette and Ayush Jain},
  journal = {arXiv (preprint, not yet peer-reviewed)},
  year = {2026},
  url = {https://arxiv.org/abs/2610.31483},
}

What it is

Preprint, not yet peer-reviewed. Researchers at Meta and Carnegie Mellon University study reinforcement learning for tool selection by a small language model and transfer to different retrievers and tool collections. The authors report that their post-trained Qwen3-8B outperforms Llama 4 Maverick on ToolBench; this is the paper’s reported result, not an independent benchmark verification.

Topics

#ai agents #tool selection #reinforcement learning #preprint

Added Oct 26, 2026 · 0 opens