The paper proposes a retrieval‑grounded credit‑assignment method for generative recommenders that use Semantic IDs (SIDs). By structuring each generated trace into a history summary, a set of interest hypotheses, and a final SID, and then verifying each hypothesis with a frozen retriever, the method assigns reward at the hypothesis level rather than only at the final SID. Experiments on Amazon Reviews datasets show consistent improvements in SID recommendation, and an oracle analysis on Video Games data demonstrates that selecting target‑relevant queries among generated interests boosts recall and ranking.
By Mengdan Zhu, Yufan Zhao, Yao Zhao, Sophie Di, Tao Di, Yulan Yan, Sridhar Iyer, Liang Zhao
arXiv:2610.22797v1 Announce Type: cross
Abstract: Reasoning rerankers write a rationale before ranking a candidate slate, and existing systems read item metadata to do so, even though generative reco...
By Zhenwei Tang, Jintao Liu, Di Bai, Xidong Wu, Xinwu Cheng
arXiv:2608. 11980v1 Announce Type: cross Abstract: Semantic-ID generative recommenders represent each item as a short sequence of discrete semantic tokens and predict the next item by autoregressively generating this token sequence.
By Kangning Zhang, Haotian Fang, Xukun Luo, Hao Yin, Yang Gao, Peng Yan, Weiwen Liu, Weinan Zhang, Yong Yu
The paper introduces Difficulty‑Aware Semantic‑ID Optimization (DASO), a post‑training method for generative recommendation that improves tree‑structured item ranking. DASO profiles rollout groups by prefix‑match depth, reallocates a portion of candidates to prefix‑guided completions, and uses a SID‑prefix reward with an auxiliary SFT anchor to address target‑missing failures. On public benchmarks, DASO outperforms MiniOneRec‑style GRPO on 11 of 12 metrics and achieves the best results on 9 of 12 metrics, also improving level‑wise recall on an internal recommendation task.
By Xin Yu, Stephen Li, Sina Aghaei, Zifan Zhu, Jiamu Bai, Guanjie Huang, Bo Peng, Yiyao Liu, Lingzhou Xue
arXiv:2610.15864v1 Announce Type: cross
Abstract: Large-scale personalization is shifting from embedding-table retrieval to LLM-based recommenders that generate the Semantic ID (SID) of the item to r...
By Sejoon Oh, Fernando Amat Gil, Evan Smothers, Dan Iter, Mark Thornburg, Dawit Mureja Argaw, Sourabh Medapati, Thea Wang, Ria Verma, Qiuling Xu, Harshit Syal, Ehsan Gholami, Moumita Bhattacharya, Ashish Rastogi
arXiv:2605. 17648v2 Announce Type: replace Abstract: Generative recommendation treats next-item prediction as autoregressive item-identifier generation.
By Zaiyi Zheng, Liang Wu, Guanghui Min, Yaochen Zhu, Liangjie Hong, Chen Chen, Jundong Li
arXiv:2608. 11980v2 Announce Type: replace-cross Abstract: Semantic-ID generative recommenders represent each item as a short sequence of discrete semantic tokens and predict the next item by autoregressively generating this token sequence.
By Kangning Zhang, Haotian Fang, Xukun Luo, Hao Yin, Yang Gao, Peng Yan, Weiwen Liu, Weinan Zhang, Yong Yu
arXiv:2610.16979v1 Announce Type: cross
Abstract: Generative retrieval (GR) offers a promising paradigm for industrial search by unifying candidate generation within an autoregressive framework. Howe...
By Jiajun Han, Guoxin Ma, Felix, Liangsheng Xu, Ruiqi Yu, Yingdi Guo, Zongbing Wang, Xinhua Ye, Chenyan Hua, Xin Xie, Pengcheng Yang, Haijun Wu
arXiv:2604. 20861v3 Announce Type: replace-cross Abstract: Semantic IDs (SIDs) provide the discrete item vocabulary used by generative recommendation, but their quality depends on what item evidence is preserved before quantization.
By Yangchen Zeng, Jinze Wang
The paper examines LLM-based recommendation rerankers that are often evaluated under an oracle protocol, which guarantees the ground-truth item is present in the scored set. Across Amazon datasets, this protocol overestimates realistic NDCG@10 by 92–95% because realistic retrieval only covers 2–19% of relevant items at K=100, creating a recall ceiling that limits any closed-candidate reranker's top‑k NDCG. The authors find that various optimisation strategies—including prompt engineering, model scaling, sequential models, supervised neural rerankers, LoRA fine‑tuning, hybrid retrieval, score‑aware prompting, and LLM+CF fusion—do not significantly improve over a collaborative‑filtering baseline under realistic retrieval, and they propose a Recall‑Aware Evaluation Protocol (RAEP) to better assess rerankers in low‑recall regimes.
By Zhaohui Wang
arXiv:2607. 24995v1 Announce Type: new Abstract: Semantic IDs (SIDs) are now a central component of generative recommendation.
By Junting Wang, Xinrui He, Yunzhe Li, Hari Sundaram
The paper introduces CAP‑Mix, a flow‑based recommender that represents items as soft mixtures of shared prototypes plus item‑specific vectors, allowing continuous models to capture shared structure without discretization. CAP‑Mix jointly trains assignments, prototypes, and item vectors, optionally leveraging pretrained Semantic ID responsibilities or full content embeddings. Across three datasets, CAP‑Mix outperforms flat lookup tables—especially on sparse Amazon datasets—and achieves significant gains in HR@5 and NDCG@5 while delivering up to 161× faster inference than discrete baselines.
By Vicente Balmaseda, Dongqi Fu, Jiarui Feng, Ren chen, Hanqing Zeng, Tianxin Wei, Yinglong Xia, Tianbao Yang, Hong Li