arXiv：cs.AI· Jinfeng Xu, Zheyu Chen, Ziyue Peng, Zheng Lin, Wenhao Yuan, Jian Chen, Shujie Li, Edith Ngai·· 7 小时前AI 评分34
推荐系统递归自我改进中的状态保留：超越后继模型精度
Beyond Successor Accuracy: State Retention for Recursive Self-Improvement in Recommendation
AI 导读
针对推荐系统递归自我改进（Rec-RSI）仅以最新模型评估每轮效果的问题，研究者提出跨代优势（CGA）来量化代际间保留的互补排序决策，并引入选择时无需标签的排序分离统计量预测应保留哪一代模型。
正文
Abstract:Recommendation recursive self-improvement (Rec-RSI) feeds recommender outputs into subsequent training. Evaluating each round solely through its latest model assumes that the successor consolidates the update, although pre- and post-update models may retain complementary ranking decisions. We term this \emph{distributed progress} and quantify it using cross-generation advantage (CGA), a marginally matched contrast between cross- and within-generation model pairs. A rank-separation statistic, label-free at selection time, predicts which family to retain. Across four datasets and three sequential recommendation encoders, the preferred retention regime varies by architecture: cross-generation pairing benefits GRU4Rec and SASRec, whereas FMLP initially favors within-generation pairing and shifts toward cross-generation pairing after a second update. Rank separation selects the stronger family in 12/12 first-update and 5/6 second-update dataset-encoder settings; on held-out tests, the selected family outperforms the direct successor in 34/36 trajectories. Five transfer mechanisms do not consistently reproduce these gains in one model. These findings establish state retention as a distinct Rec-RSI problem: progress may reside in relations between generations as well as in the latest model. Code is available at \href{this https URL}{this https URL}.
| Subjects: | Information Retrieval (cs.IR); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2610.07105 [cs.IR] |
| (or arXiv:2610.07105v1 [cs.IR] for this version) | |
| https://doi.org/10.48550/arXiv.2610.07105 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Jinfeng Xu [view email]
[v1]
Mon, 5 Oct 2026 15:03:27 UTC (141 KB)
来源：arXiv：cs.AI · arxiv.org