8 papers
BEAM: Binary Expert Activation Masking for Dynamic Routing in MoE
Juntong Wu, Jialiang Cheng, Qishen Yin +5
Mixture-of-Experts (MoE) architectures enhance the efficiency of large language models by activating only a subset of experts per token. However, standard MoE employs a fixed Top-K…
Synthetic Data Powers Product Retrieval for Long-tail Knowledge-Intensive Queries in E-commerce Search
Gui Ling, Weiyuan Li, Yue Jiang +6
Product retrieval is the backbone of e-commerce search: for each user query, it identifies a high-recall candidate set from billions of items, laying the foundation for high-qualit…
SERE: Similarity-based Expert Re-routing for Efficient Batch Decoding in MoE Models
Juntong Wu, Jialiang Cheng, Fuyu Lv +2
Mixture-of-Experts (MoE) architectures employ sparse activation to deliver faster training and inference with higher accuracy than dense LLMs. However, in production serving, MoE m…
Retrieval-GRPO: A Multi-Objective Reinforcement Learning Framework for Dense Retrieval in Taobao Search
Xingxian Liu, Dongshuai Li, Jiahui Wan +7
Dense retrieval, as the core component of e-commerce search engines, maps user queries and items into a unified semantic space through pre-trained embedding models to enable large-…
C2T-ID: Converting Semantic Codebooks to Textual Document Identifiers for Generative Search
Yingchen Zhang, Ruqing Zhang, Jiafeng Guo +4
Designing document identifiers (docids) that carry rich semantic information while maintaining tractable search spaces is a important challenge in generative retrieval (GR). Popula…
LLMs as Sparse Retrievers:A Framework for First-Stage Product Search
Hongru Song, Yu-an Liu, Ruqing Zhang +6
Product search is a crucial component of modern e-commerce platforms, with billions of user queries every day. In product search systems, first-stage retrieval should achieve high…