7 citations
- Aarhus UniversityDK1 paper
- Amazon (Germany)DE1 paper
- Amazon (United Kingdom)GB1 paper
- Boston Dynamics (United States)US1 paper
- Grand Valley State UniversityUS1 paper
- ILC Dover (United States)US1 paper
- Johns Hopkins UniversityUS1 paper
- KU LeuvenBE1 paper
- Massachusetts Institute of TechnologyUS1 paper
- Queen Mary University of LondonGB1 paper
- RMIT UniversityAU1 paper
- Samsung (South Korea)KR1 paper
4 papers · 1 filter
Layer-wise Token Compression for Efficient Document Reranking
Shengyao Zhuang, Zhichao Xu, Ivano Lauriola
Transformer-based document cross-encoder rerankers are a central component of modern information retrieval systems. Despite their success, these models suffer from high computation…
Multi-modal Relational Item Representation Learning for Inferring Substitutable and Complementary Items
Junting Wang, Chenghuan Guo, Jiao Yang +3
We study the problem of inferring substitutable and complementary items, which underpins applications such as alternative and follow-up purchase suggestions. Existing approaches ty…
From Unstructured to Structured: LLM-Guided Attribute Graphs for Entity Search and Ranking
Yilun Zhu, Nikhita Vedula, Shervin Malmasi
Entity search, i.e., finding the most similar entities to a query entity, faces unique challenges in e-commerce, where product similarity varies across categories and contexts. Tra…
Why Advanced Encoders Lag on Sparse Retrieval? The Answer and an Approach to Bridging Vocabulary Gaps
Zhichao Geng, Yang Yang
While advanced foundation models like ModernBERT significantly outperform older architectures in dense retrieval, they surprisingly lag behind the aging BERT-base baseline in learn…