Metraipolis › archive › dc_e4e2431e62a3
History-Conditioned Joint-Prefix Alignment for Generative Recommendation — arXiv:2608.29179
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2608.29179
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-17T07:47:32.259Z
- SHA-256 of the archived text
68e7c92010bbf2892ca7ab47d7a3faff15b997fb3abc12692f0b74f2a8f59355- Stable URL of this page
https://metraipolis.com/documento/dc_e4e2431e62a3
How to cite
History-Conditioned Joint-Prefix Alignment for Generative Recommendation — arXiv:2608.29179 — arXiv (en), https://arxiv.org/abs/2608.29179 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_e4e2431e62a3 (Metraipolis archive dc_e4e2431e62a3, sha256 68e7c92010bbf2892ca7ab47d7a3faff15b997fb3abc12692f0b74f2a8f59355)
BibTeX
@misc{dce4e2431e62a3,
title = {History-Conditioned Joint-Prefix Alignment for Generative Recommendation — arXiv:2608.29179},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2608.29179},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_e4e2431e62a3; sha256 68e7c92010bbf2892ca7ab47d7a3faff15b997fb3abc12692f0b74f2a8f59355},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.29179. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Generative recommendation retrieves items by autoregressively generating semantic identifiers, but beam search may discard a target before its complete identifier is generated. Our preliminary analysis across three benchmarks shows that most missed targets are pruned within the first two decoding steps, highlighting the importance of early prefix retention. However, retaining a target's first-token branch alone is insufficient if its continuation is pruned at the next step: aligning only the first-token distribution improves first-token survival but yields little gain in complete-path retention. This observation motivates joint supervision of early branches and their continuations. We propose Prefix Alignment with Temporal History (PATH), which aggregates transition statistics from the training corpus over recent interactions with exponential decay to construct history-conditioned two-token prefix targets. PATH aligns the model's joint predictions with these targets through forward KL divergence, using a chain-rule decomposition that enables unbiased Monte Carlo estimation. At inference time, PATH reuses the transition statistics for pointwise mutual information (PMI) calibration, reranking completed candidates relative to global prefix frequency. Experiments on Beauty, Instruments, and Yelp demonstrate the effectiveness of PATH, showing consistent improvements in recommendation performance and higher full-SID survival rates. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-01 in cs.AI). Categorie arXiv: Information Retrieval (cs.IR) ; Artificial Intelligence (cs.AI). Pagina del lavoro: https://arxiv.org/abs/2608.29179 — 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 2026-10-17T07:28:26.725Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-01?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).