Probabilistic Kernel Function for Fast Angle Testing

ICLROral2026

Authors
Kejing Lu, Chuan Xiao, Yoshiharu Ishikawa
Affiliation
Yamanashi University
Venue
ICLR 2026
Track
Oral

TL;DR

We propose two probabilistic kernel functions for angle testing, which can be used to accelerate vector-based similarity search.

Opening excerpt from the authors’ abstract. source

Read the paper

Topics

kernel

← All ICLR 2026 Oral papers · Browse the whole archive