Metraipolis › archive › dc_0377be82df5b
Small Agents with Semantic Search: Efficient Multilingual Code Localization — arXiv:2610.05099
- Source
- arXiv
- Original
- https://arxiv.org/abs/2610.05099
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-06T05:28:15.317Z
- SHA-256 of the archived text
ecff83c947045b81054a1566b82decea8418a69eb1581c8cc49e91ecef6f7ffa- Stable URL of this page
https://metraipolis.com/documento/dc_0377be82df5b
How to cite
Small Agents with Semantic Search: Efficient Multilingual Code Localization — arXiv:2610.05099 — arXiv, https://arxiv.org/abs/2610.05099 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_0377be82df5b (Metraipolis archive dc_0377be82df5b, sha256 ecff83c947045b81054a1566b82decea8418a69eb1581c8cc49e91ecef6f7ffa)
BibTeX
@misc{dc0377be82df5b,
title = {Small Agents with Semantic Search: Efficient Multilingual Code Localization — arXiv:2610.05099},
howpublished = {arXiv},
url = {https://arxiv.org/abs/2610.05099},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_0377be82df5b; sha256 ecff83c947045b81054a1566b82decea8418a69eb1581c8cc49e91ecef6f7ffa},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv, https://arxiv.org/abs/2610.05099. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Locating relevant files from natural-language requests is a core subtask for agents operating over code repositories. We investigate whether this task can be delegated to compact, specialized models to enable on-device search while reducing the token usage, latency, and inference cost of larger agents. We show that semantic search improves file localization, with gains in accuracy, cross-language transfer, and inference efficiency. To study this setting, we introduce a training framework for file-localization agents built around ColGREP, a local semantic search tool based on late-interaction retrieval models. Our recipe combines weighted supervised fine-tuning on teacher trajectories, assigning turn-level credit based on retrieval outcomes, with reinforcement learning on localization quality. We train three model families with fewer than two billion parameters to formulate search queries, inspect retrieved content, and identify relevant files. On localization tasks derived from SWE-bench Lite and Multi-SWE-bench Flash, ColGREP-equipped agents substantially improve over their base models and outperform corresponding GREP-based agents. In addition to improving localization accuracy, ColGREP reduces mean end-to-end trajectory latency by 44.1\% on CPU while using 29.1\% fewer tokens, and enables better generalization to programming languages unseen during fine-tuning. These results suggest that compact, tool-specialized localization agents can provide an efficient interface between natural-language requests and large codebases. Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.CL, nuovi lavori del giorno). Categorie arXiv: Computation and Language (cs.CL). Pagina del lavoro: https://arxiv.org/abs/2610.05099 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-10-06T05:24:20.306Z dalla pagina https://arxiv.org/list/cs.CL/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).