Metraipolis › archive › dc_da0cfc92f6ff

Towards Efficient Reasoning in LLM-Based Recommender Systems via Model Merging — arXiv:2608.10447

Source
arXiv (en)
Original
https://arxiv.org/abs/2608.10447
License (as archived)
CC0-1.0
Archived
2026-10-19T06:09:10.657Z
SHA-256 of the archived text
195935d05e45176b8e50d72bbfafbb2b6783cf6897825bb563f27214c10f4a18
Stable URL of this page
https://metraipolis.com/documento/dc_da0cfc92f6ff

How to cite

Towards Efficient Reasoning in LLM-Based Recommender Systems via Model Merging — arXiv:2608.10447 — arXiv (en), https://arxiv.org/abs/2608.10447 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_da0cfc92f6ff (Metraipolis archive dc_da0cfc92f6ff, sha256 195935d05e45176b8e50d72bbfafbb2b6783cf6897825bb563f27214c10f4a18)
BibTeX
@misc{dcda0cfc92f6ff,
  title = {Towards Efficient Reasoning in LLM-Based Recommender Systems via Model Merging — arXiv:2608.10447},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2608.10447},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_da0cfc92f6ff; sha256 195935d05e45176b8e50d72bbfafbb2b6783cf6897825bb563f27214c10f4a18},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.10447. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Large language model-based recommender systems are increasingly adopting slow-thinking models that generate step-by-step reasoning before making predictions, often achieving higher accuracy than fast-thinking models that predict directly. However, their reasoning traces are often unnecessarily verbose, increasing inference costs without commensurate accuracy gains. Existing training-based approaches to reasoning compression often incur substantial adaptation costs, while inference-time methods are brittle and difficult to scale. These limitations motivate model merging as a promising training-free direction for transferring specialised behaviours between models in a shared parameter space. In particular, merging a slow-thinking model with a fast-thinking counterpart provides a natural mechanism for balancing recommendation accuracy and reasoning conciseness. To this end, we propose, to our knowledge, the first model merging framework for reasoning compression in recommender systems. Unlike conventional merging methods that apply uniform merge coefficients across model components, our method performs fine-grained merging at the level of individual attention heads, capturing heterogeneous patterns in recommendation reasoning. Each attention head is assigned a distinct merge coefficient according to its contribution to critical reasoning evidence and its sensitivity to parameter change, enabling selective injection of the concise behaviour of the fast-thinking model into the slow-thinking model and reducing reasoning verbosity without compromising recommendation quality. Experiments on three benchmark datasets show that our method reduces reasoning length by up to 24.3% while outperforming competitive model merging baselines in maintaining recommendation accuracy. The code is available at this https URL .

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-12 in cs.AI).
Categorie arXiv: Information Retrieval (cs.IR) ; Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2608.10447 — 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 2026-10-19T05:57:28.368Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-08-12?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).