Metraipolis › archive › dc_4322b5876667

Do Recommenders Need to Think in Words? Latent Reasoning for Efficient LLM-Based Recommendation — arXiv:2611.07866

Source
arXiv (en)
Original
https://arxiv.org/abs/2611.07866
License (as archived)
CC0-1.0
Archived
2026-11-06T14:43:11.542Z
SHA-256 of the archived text
498ec1062b4319b9d2981761cc0f9f100749b73c2bbc501e77344513a78bcb56
Stable URL of this page
https://metraipolis.com/documento/dc_4322b5876667

How to cite

Do Recommenders Need to Think in Words? Latent Reasoning for Efficient LLM-Based Recommendation — arXiv:2611.07866 — arXiv (en), https://arxiv.org/abs/2611.07866 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_4322b5876667 (Metraipolis archive dc_4322b5876667, sha256 498ec1062b4319b9d2981761cc0f9f100749b73c2bbc501e77344513a78bcb56)
BibTeX
@misc{dc4322b5876667,
  title = {Do Recommenders Need to Think in Words? Latent Reasoning for Efficient LLM-Based Recommendation — arXiv:2611.07866},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2611.07866},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_4322b5876667; sha256 498ec1062b4319b9d2981761cc0f9f100749b73c2bbc501e77344513a78bcb56},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.07866. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

LLM-based recommender systems increasingly generate chain-of-thought (CoT) text before each item prediction to activate the LLM's reasoning capacity, costing tens to hundreds of extra tokens per recommendation at inference. We ask whether inference-time reasoning artifacts (textual or latent) can be replaced by training-time supervision alone, without loss of accuracy. We propose LatentRec, a framework for LLM-based generative recommendation that replaces inference-time reasoning artifacts with training-time per-step head supervision. LatentRec has three components. A 2.5M-parameter Continuous Thought Module iterates K latent reasoning steps over the LLM's hidden state during training. Soft Token Selection bridges the ~14x norm gap between hidden states and input embeddings. A Per-step Reasoning Loss (PRL) supervises every intermediate state through the LLM's prediction head against the first token of the target Semantic ID. The trained LLM retains this accuracy without the module at inference, so the module can be discarded at deployment. On three Amazon benchmarks, LatentRec improves HR@10 by 4.0-5.0% (relative) over a CoT-distilled baseline at 18.3x lower latency (batch=1). Per-step training supervision consistently improves accuracy across three datasets, paid once at training rather than at every recommendation.

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/2611.07866 — 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 2026-11-06T14:18:57.589Z dalla pagina https://arxiv.org/list/cs.CL/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).