Metraipolis › archive › dc_ec37b48999cd

Distill Globally, Adapt Locally: Reasoning Distillation and Product-Type Test-Time Training for Scalable Trade-Up Recommendation — arXiv:2609.05363

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.05363
License (as archived)
CC0-1.0
Archived
2026-10-16T22:59:57.398Z
SHA-256 of the archived text
f807857b33d2bde19a866d37af111eadc240cb54a9b230e64567d52fa40c3555
Stable URL of this page
https://metraipolis.com/documento/dc_ec37b48999cd

How to cite

Distill Globally, Adapt Locally: Reasoning Distillation and Product-Type Test-Time Training for Scalable Trade-Up Recommendation — arXiv:2609.05363 — arXiv (en), https://arxiv.org/abs/2609.05363 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_ec37b48999cd (Metraipolis archive dc_ec37b48999cd, sha256 f807857b33d2bde19a866d37af111eadc240cb54a9b230e64567d52fa40c3555)
BibTeX
@misc{dcec37b48999cd,
  title = {Distill Globally, Adapt Locally: Reasoning Distillation and Product-Type Test-Time Training for Scalable Trade-Up Recommendation — arXiv:2609.05363},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.05363},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_ec37b48999cd; sha256 f807857b33d2bde19a866d37af111eadc240cb54a9b230e64567d52fa40c3555},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.05363. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Trade-up recommendation identifies higher-quality alternatives that preserve a customer's purchase intent while offering upgraded benefits. Large language models (LLMs) can reason about such distinctions, but applying them directly to hundreds of millions of product pairs is operationally impractical. We introduce a two-level framework that distills LLM reasoning into an efficient non-generative student and adapts its decision boundary to product-type-specific trade-up criteria. At Level 1, a retrieval-augmented few-shot LLM teacher generates structured relation labels and natural-language rationales. These rationales supervise a compact embedding-pair classifier through alignment and contrastive objectives; at inference, the student uses only two precomputed 768-dimensional product embeddings, with no LLM calls or text generation. On a fixed human-annotated benchmark of 8,352 pairs, a 15.5M-parameter four-class reasoning-distilled student achieves AUC 0.924 (95% CI [0.918, 0.929]), compared with 0.912 for the four-class label-only student. At Level 2, product-type test-time training (PT-TTT) uses few-shot demonstrations to optimize lightweight category-specific adapters over the frozen student. PT-TTT improves AUC from 0.924 to 0.941 and average precision from 0.920 to 0.940. On a 100K-pair proxy catalog, the distilled student on a single eight-GPU machine is approximately 5,000x faster and 10,000x lower in estimated cost than direct LLM inference.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-07 in cs.LG).
Categorie arXiv: Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2609.05363 — 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 2026-10-16T22:31:32.219Z dalla pagina https://arxiv.org/catchup/cs.LG/2026-09-07?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).