AI Paper Atlas

Research map / Architectures and efficiency / RNNs and normalization

LEMUR: Learned Multi-Vector Retrieval

Elias Jääsaari, Ville Hyvönen, Teemu Roos

ICML 2026 · Architectures and efficiency › RNNs and normalization · approach: Learned embeddings for retrieval · 0 citations (OpenAlex)

Read the paper

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
use LEMUR learned single-vector reduction instead of best baseline (MUVERA/IGP/DESSERT/PLAID)multi-vector similarity search · BEIR datasets, ColBERTv2, k=100QPS at >=80% recall: 5-16x faster than the best-performing baselineImproved
increase LEMUR hidden layer size d' (1024, 2048, 4096) (ablation)LEMUR · BEIR, Recall100@k'recallImproved
increase d' from 2048 to 4096 (ablation)LEMUR · BEIR end-to-end latencyend-to-end latencyNo clear effect
increase d' from 1024 to 2048 (ablation)LEMUR · BEIR end-to-end latencylatencyImproved
use 1024-dim LEMUR embeddings instead of 10240-dim MUVERA FDEs (ablation)single-vector candidate retrieval · BEIR, Recall100@k' (seven of eight datasets)recallImproved
increase MLP depth beyond two layers (ablation)LEMUR · MaxSim estimationperformanceNo clear effect
use random (untrained) hidden-layer weights instead of trained feature encoder (ablation)LEMUR · training ablation (Appendix H.3)retrieval performanceHurt
train on actual queries instead of corpus documents encoded with the query encoder (ablation)LEMUR · training set selectionretrieval performanceImproved
use LEMUR instead of MUVERA on non-ColBERTv2 embeddingsmulti-vector search · Quora, answerai-colbert-small, GTE-ModernColBERT, LFM2-ColBERTrecall: MUVERA fails to exceed 60% recallImproved
use LEMUR on visual document retrieval modelsmulti-vector search · ViDoRe, ColModernVBERT and ColQwen2QPS vs recallImproved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy