Semantic-ID-based generative recommendation unifies retrieval and ranking through autoregressive decoding but differs structurally from discriminative models, which can directly access rich target-item side features. The attributes are compressed into a finite codebook during quantization, while the target item remains...
Jun Yang, Jia-Siang Chen, Jia-Wei Sun et al.· 0 citations
E-commerce search often suffers from vocabulary mismatch between user queries and merchant-authored product titles, since short titles cannot fully cover diverse user expressions or visual product attributes. Although Doc2Query alleviates this issue by generating pseudo-queries for document expansion, traditional metho...
MIMA is proposed, a Multi-Interest recommendation framework built on Multi-positive exclusive Assignment that groups items co-occurring within the same request into a positive set, generates complementary interests with a causal Transformer decoder, and exclusively assigns each positive to supervise a distinct interest...
Xing-Yuan Mao, Alin Fan, Shi-Chao Nie et al.· 0 citations
A transferable residual adapter that injects additional ranking-specific features into ranking in a residual manner and an asymmetric multi-epoch training strategy that resets sparse parameters while continuously accumulating dense parameters across epochs, alleviating the overfitting of sparse parameters are proposed.
Xiao-Dong Li, Alin Fan, Ming-Yang Li et al.· 0 citations
This work proposes EAGER (Enrich-and-AliGn gEnerative Query Recommendation), a two-stage framework for generating query suggestions from clicked items that has been deployed in production at a major e-commerce platform.
Shu-Wei Yuan, Mingyu Ding, Lu-Xin Liu et al.· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
