研究人员开发了多种优化物品分词以改进生成式推荐系统的新方法。其中一种方法Tlow使用基于流的模型将语义嵌入转换为标准正态分布，从而实现独立分词，并将微信等平台的用户点击率提高了10%以上。另一种方法DASO通过优化基于难度的奖励来解决基于语义ID的生成中的挑战，在多个基准测试中表现优于现有的GRPO方法。此外，一种动态单层语义码本方法降低了自回归解码成本并提高了检索指标，在在线A/B测试中主要消费指标提高了0.792%。另一种技术语义子词分词（SST）使用可变长度的语义子词来减少物品内注意力过载并改进物品间行为建模。
AI
Applying scaling laws to recommendation retrieval is hindered by feature heterogeneity: naively stacking Transformer layers yields diminishing returns because heterogeneous fields produce severe token-norm divergence. We present TransRetrieval, a Transformer-based retrieval frame…
arXiv:2608.24176v1 Announce Type: cross Abstract: Item tokenizer encodes semantic embeddings into token IDs to replace the randomly assigned item IDs used in traditional recommendation models, fundamentally addressing the problems of excessive parameters and cold starts. However,…
Item tokenizer encodes semantic embeddings into token IDs to replace the randomly assigned item IDs used in traditional recommendation models, fundamentally addressing the problems of excessive parameters and cold starts. However, the most common tokenizer, RQ-VAE, suffers from l…
arXiv cs.AI
TIER_1English(EN)·Xin Yu, Stephen Li, Sina Aghaei, Zifan Zhu, Jiamu Bai, Guanjie Huang, Bo Peng, Yiyao Liu, Lingzhou Xue·
arXiv:2608.20611v1 Announce Type: new Abstract: Semantic-ID-based generative recommendation casts retrieval and ranking as autoregressive generation over hierarchical item identifiers. A common recipe is SFT followed by GRPO, yet vanilla GRPO is poorly matched to this tree-struct…
arXiv:2608.21012v1 Announce Type: cross Abstract: Generative recommendation represents each item with a sequence of discrete Semantic IDs (SIDs) and predicts the sequence to retrieve the next item. Typical systems use multi-level residual quantization, which increases autoregress…
In generative recommender systems, items are typically tokenized into fixed-length semantic ID sequences for autoregressive next-item prediction. However, for user-context modeling, this fine-grained representation triggers Intra-item Attention Overload: excessive attention is sp…
Generative recommendation represents each item with a sequence of discrete Semantic IDs (SIDs) and predicts the sequence to retrieve the next item. Typical systems use multi-level residual quantization, which increases autoregressive decoding cost and creates a large hierarchical…
