AI RESEARCH
Coordinate Heterogeneity Governs Binary Quantization: From InfoNCE to Recall
arXiv CS.LG
•
ArXi:2605.17524v1 Announce Type: new Binary quantization (BQ) compresses high-dimensional embeddings into one or two bits per coordinate, enabling nearest neighbor search at extreme speed. Yet a striking puzzle persists: BQ achieves competitive recall on contrastive embeddings but fails on others -- and two leading systems adopt diametrically opposite strategies (random rotation vs. preserving coordinate axes) without a common theory explaining when each is appropriate.