Metraipolis › archive › dc_c3e5207fd58e
When Memory Takes Gradients: Collaborative Vector Memory for Agentic Recommender Systems — arXiv:2608.26895
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2608.26895
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-17T14:30:17.935Z
- SHA-256 of the archived text
185a475191d83b6e49f7441fd7efc6aaa0a9c31598aba59db1246f66a5a1117a- Stable URL of this page
https://metraipolis.com/documento/dc_c3e5207fd58e
How to cite
When Memory Takes Gradients: Collaborative Vector Memory for Agentic Recommender Systems — arXiv:2608.26895 — arXiv (en), https://arxiv.org/abs/2608.26895 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_c3e5207fd58e (Metraipolis archive dc_c3e5207fd58e, sha256 185a475191d83b6e49f7441fd7efc6aaa0a9c31598aba59db1246f66a5a1117a)
BibTeX
@misc{dcc3e5207fd58e,
title = {When Memory Takes Gradients: Collaborative Vector Memory for Agentic Recommender Systems — arXiv:2608.26895},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2608.26895},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_c3e5207fd58e; sha256 185a475191d83b6e49f7441fd7efc6aaa0a9c31598aba59db1246f66a5a1117a},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.26895. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Agentic recommender systems ground each decision of a large language model (LLM) in a persistent memory of the user, and in existing agents that memory is text: a narrative written and maintained by further LLM calls. Text limits this memory in two ways. It is updated one rewrite at a time, so exploiting the full interaction history is prohibitively expensive; and collaborative evidence, graded similarity over an entire catalog, does not survive translation into sentences. We propose CoVeMem (Collaborative Vector Memory), which vectorizes the collaborative core of the agent's memory. Frozen LightGCN user and item states form the memory bank; at each decision, the candidate set itself retrieves the most relevant historical states, which enter the LLM's context as soft tokens alongside a light textual profile. Contrastive alignment to item-semantic anchors, followed by listwise co-training with masked candidates, teaches the model to read these states and to rank through them; a pointwise yes/no readout scores each candidate. Across four instruction-grounded recommendation benchmarks, CoVeMem matches or exceeds the strongest collaborative text-memory agent on 19 of 20 metric cells while requiring zero additional LLM calls for memory maintenance beyond the shared static profile, against per-interaction calls for text memory. The memory now takes gradients: the full interaction history, out of reach for text, becomes available as training data for what the agent remembers and for how it reads what it remembers. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-28 in cs.AI). Categorie arXiv: Information Retrieval (cs.IR) ; Artificial Intelligence (cs.AI). Pagina del lavoro: https://arxiv.org/abs/2608.26895 — 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 2026-10-17T13:16:53.675Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-08-28?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).