BEIR benchmark
PulseAugur coverage of BEIR benchmark — every cluster mentioning BEIR benchmark across labs, papers, and developer communities, ranked by signal.
-
New framework bridges vocabulary gap to boost AI sparse retrieval performance
Researchers have identified a "vocabulary gap" as the reason why advanced foundation models like ModernBERT underperform older models in learned sparse retrieval tasks. This gap arises because modern tokenizers use raw,…
-
Apple ML Research proposes amortized MIPS using neural networks
Apple Machine Learning Research has published a paper detailing a new approach to Maximum Inner Product Search (MIPS) called amortized MIPS. This method utilizes neural networks, specifically SupportNet and KeyNet, to d…
-
New Amortized MIPS Approach Uses Neural Networks for Faster Search
Researchers have developed a novel approach called amortized maximum inner product search (MIPS) that utilizes neural networks to directly predict MIPS solutions. This method trains networks to act as "support functions…
-
New retrieval method replaces K-means with sparse coding for faster, more accurate results
Researchers have introduced Single-stage Sparse Retrieval (SSR), a new method for efficient multi-vector retrieval that bypasses traditional K-means clustering. SSR utilizes Sparse Autoencoders to create high-dimensiona…