Showing cs.IRShow all
3 papers · 1 filter
cs.IR2026
A Brief Comparison of Training-Free Multi-Vector Sequence Compression Methods
Rohan Jha, Chunsheng Zuo, Reno Kriz +1
While multi-vector retrieval models outperform single-vector models of comparable size in retrieval quality, their practicality is limited by substantially larger index sizes, driv…
cs.IR2026
Multi-Vector Index Compression in Any Modality
Hanxiang Qin, Alexander Martin, Rohan Jha +3
We study efficient multi-vector retrieval for late interaction in any modality. Late interaction has emerged as a dominant paradigm for information retrieval in text, images, visua…
cs.IR2026
More Than Efficiency: Embedding Compression Improves Domain Adaptation in Dense Retrieval
Chunsheng Zuo, Daniel Khashabi
Dense retrievers powered by pretrained embeddings are widely used for document retrieval but struggle in specialized domains due to the mismatches between the training and target d…