Metraipolis › archive › dc_484221d524dd

On-Device Commercial Intent Retrieval Under Size, Latency, and Privacy Constraints: A 3 MiB Retrieval System with Typed Egress Boundaries — arXiv:2610.00170

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.00170
License (as archived)
CC0-1.0
Archived
2026-10-06T13:56:48.568Z
SHA-256 of the archived text
e4a8cda355925fbb697edcd688dcbecb93bf08919c884ff09cdb31492960c334
Stable URL of this page
https://metraipolis.com/documento/dc_484221d524dd

How to cite

On-Device Commercial Intent Retrieval Under Size, Latency, and Privacy Constraints: A 3 MiB Retrieval System with Typed Egress Boundaries — arXiv:2610.00170 — arXiv (en), https://arxiv.org/abs/2610.00170 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_484221d524dd (Metraipolis archive dc_484221d524dd, sha256 e4a8cda355925fbb697edcd688dcbecb93bf08919c884ff09cdb31492960c334)
BibTeX
@misc{dc484221d524dd,
  title = {On-Device Commercial Intent Retrieval Under Size, Latency, and Privacy Constraints: A 3 MiB Retrieval System with Typed Egress Boundaries — arXiv:2610.00170},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.00170},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_484221d524dd; sha256 e4a8cda355925fbb697edcd688dcbecb93bf08919c884ff09cdb31492960c334},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.00170. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

We study commercial intent inference that runs entirely on the user's device, under three constraints frozen before the work began: the downloaded payload under 3 MiB, Tier-0 inference under 20 ms at p95, and no raw text, content embedding, or stable identifier leaving the device. Under them we build a retrieval path over a 6,020-leaf commercial taxonomy: a static embedding table distilled from a Korean sentence transformer, quantized to 4 bits, no inference runtime. Our main result is where that constraint costs accuracy. On real Korean commerce text labelled by others (22,900 AI-Hub shopping reviews), mid-category top-5 on real product names is 75.0% against an 18.4% permutation baseline, but splits on one observable: a query containing some leaf name as a substring scores 83.5%, one containing none 45.2%. A generic 196.6x larger teacher seemed to localize the gap (+20.1 pp without an anchor, +0.1 with). That null was two effects cancelling: the same teacher fine-tuned on the student's own contrastive pairs reaches 0.8586 and beats the pure-encoder student by +10.6 pp with an anchor and +20.9 pp without. The cost is not uniform, but it is not free anywhere; where the anchor is absent, task adaptation buys the teacher nothing, so what the constrained encoder lacks there is capacity. The expensive regime is detectable on-device from the ranker's own score margin: declining the least confident fifth lifts the rest to 0.8296. A second axis we first reported, a manufacturer model code, does not survive source-category fixed effects (-4.0 pp, p=0.51); the anchor does (+13.0 pp). Payload is 2,942,652 bytes, all three library links measured. Tier-0 p95 is 4.431 and 3.670 ms on two iPhones (A14, A16) and 5.080 ms on a budget Android tablet (Snapdragon 695), all slower than three server CPUs on the same code. Taxonomy supervision is mostly synthetic Korean utterances.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-10-02 in cs.CL).
Categorie arXiv: Information Retrieval (cs.IR) ; Computation and Language (cs.CL).
Pagina del lavoro: https://arxiv.org/abs/2610.00170 — 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 2026-10-06T12:15:34.794Z dalla pagina https://arxiv.org/catchup/cs.CL/2026-10-02?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).