{"ID":6620447,"CreatedAt":"2026-07-15T01:01:48.440468303Z","UpdatedAt":"2026-07-15T03:28:55.185153975Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2607.12281","arxiv_id":"2607.12281","title":"SlimPer: Make Personalization Model Slim and Smart","abstract":"Transformer-style architectures are increasingly adopted for industrial recommendation systems, yet they inherit a design premise misaligned with the task: generative models rely on per-token autoregressive prediction, which justifies maintaining large intermediate tensors that scale with sequence length. In contrast, recommendation systems produce a single set of relevance scores for each \u003cuser, item\u003e pair without token-level supervision. Leveraging this observation, we propose SlimPer, which reformulates personalized ranking as iterative refinement of a compact, unified \u003cuser, item\u003e knowledge base. At each layer, the model selectively queries raw multi-modal user-side tokens, computes explicit relevance matching scores, and refines the knowledge base, all in O(N) per-layer cost with a fixed-size intermediate representation. As a result, model depth is decoupled from user history length, enabling deeper relevance understanding without proportional growth in compute or memory; request-only optimization further trims memory by sharing a single copy of user-side tokens across all candidate items. SlimPer unifies sparse, dense, and sequence features within a single backbone and provides inherent interpretability through its attention mechanism. Deployed on Instagram Reels and Feed, SlimPer yields measurable improvements in user engagement while streamlining the overall system and enabling effective modeling of 10k+ fine-grained user history events.","short_abstract":"Transformer-style architectures are increasingly adopted for industrial recommendation systems, yet they inherit a design premise misaligned with the task: generative models rely on per-token autoregressive prediction, which justifies maintaining large intermediate tensors that scale with sequence length. In contrast,...","url_abs":"https://arxiv.org/abs/2607.12281","url_pdf":"https://arxiv.org/pdf/2607.12281v1","authors":"[\"Siqi Wang\",\"Xianjie Chen\",\"Shaofeng Deng\",\"Albert Chen\",\"Romil Shah\",\"Jiawei Huang\",\"Zhaoqin Wang\",\"Zhang Zhang\",\"Yiqun Liu\",\"Meilei Jiang\",\"Anish Dubey\",\"Moyan Mei\",\"Tongxin Wang\",\"Nathan Berrebbi\",\"Misael Manjarres\",\"Armand Sauzay\",\"Shardul Kothapalli\",\"Aryaman Vinchhi\",\"Kevin Johnstone\",\"Juheon Lee\",\"Gufan Yin\",\"Ziheng Huang\",\"Justin Lin\",\"Mert Terzihan\",\"Yilin Qi\",\"Cynthia Yang\",\"Colin Peppler\",\"Qi Ding\",\"Ruohan Sun\",\"Ge Song\",\"Litao Deng\",\"Parichay Kapoor\",\"Matt Ma\",\"Huihui Cheng\",\"Jiyuan Zhang\",\"Yanli Zhao\",\"Yiping Han\",\"Fangqiu Han\",\"Ning Yao\",\"Arun Singh\",\"Jordan Edwards\",\"Zhengyu Su\",\"Abhishek Kumar\",\"Guangdeng Liao\",\"Ankit Asthana\"]","published":"2026-07-14T02:31:32Z","proceeding":"cs.IR","tasks":"[\"cs.IR\",\"cs.LG\"]","methods":"[\"Transformer\"]","has_code":false}