Metraipolis › archive › dc_12cac8b27740
Selection Is Retrieval, Abstention Is Not: On-Device Tool Routing over 70 Korean-English Actions — arXiv:2609.18672
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.18672
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-15T13:49:32.372Z
- SHA-256 of the archived text
318fee18f3b77f37aaa6f29e65a0779cd0a1786481602503c213ca7b5f534a82- Stable URL of this page
https://metraipolis.com/documento/dc_12cac8b27740
How to cite
Selection Is Retrieval, Abstention Is Not: On-Device Tool Routing over 70 Korean-English Actions — arXiv:2609.18672 — arXiv (en), https://arxiv.org/abs/2609.18672 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_12cac8b27740 (Metraipolis archive dc_12cac8b27740, sha256 318fee18f3b77f37aaa6f29e65a0779cd0a1786481602503c213ca7b5f534a82)
BibTeX
@misc{dc12cac8b27740,
title = {Selection Is Retrieval, Abstention Is Not: On-Device Tool Routing over 70 Korean-English Actions — arXiv:2609.18672},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.18672},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_12cac8b27740; sha256 318fee18f3b77f37aaa6f29e65a0779cd0a1786481602503c213ca7b5f534a82},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.18672. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
An AI assistant that calls tools makes two decisions on every request: which tool to invoke, and whether any available tool applies. In the usual design a single language model makes both, by emitting a call or by declining to emit one. On a device that has to answer without a server, the language model is what makes that design expensive, dominating both the latency and the memory of the router. The common alternative is to remove the model completely and rank the catalog of local actions with a retriever instead. That substitution is not symmetric across the two decisions. A retriever returns its highest-scoring candidate for every input and cannot signal that the catalog holds no valid action. Our earlier study found that constraining a decoder to a tool grammar repairs malformed output without improving the choice. What the substitution costs in each decision has not been measured. We evaluate the two decisions separately over 600 Korean and English requests and a catalog of 70 local actions. The router may also ask for a missing slot, reply, or delegate. Half the in-catalog requests reuse catalog vocabulary and half paraphrase it, separating lexical overlap from the action requested. Character 3-gram BM25 selects 162 of 164 lexically matched requests and 85 of 166 paraphrases. Restricting the candidate set to seven raises the paraphrase figure to a mean of 0.825 over five trials. No classifier over its score features separates in-catalog from out-of-catalog above 0.697 area under the curve, where the frozen encoder multilingual-e5-base reaches 0.806. Using that encoder for abstention alone keeps 376 of the requests local and misroutes 9 of the 150 needing delegation. Abstention, not selection, is where a neural component is required. A neural ranker improves every quality metric and is rejected on latency and memory rather than accuracy. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-17 in cs.CL). Categorie arXiv: Computation and Language (cs.CL). Pagina del lavoro: https://arxiv.org/abs/2609.18672 — 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 2026-10-15T13:48:24.335Z dalla pagina https://arxiv.org/catchup/cs.CL/2026-09-17?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).