LodeHQ
AI news for people who shipFree, by email

AI · huggingface.co · Sunday, 15 November 2026

Token pooling keeps 97.8% of retrieval quality at one-quarter storage

A pooling-aware training method lets multi-vector search models cut document vectors 4x while retaining 97.8% of their unpooled NanoBEIR nDCG@10, up from 94.6% with normal training. It teaches the encoder what information will survive compression, then merges similar token vectors with hierarchical clustering. In the post’s example, pooling 200 vectors to 50 cuts quantized storage for one million documents from 12.8 GB to 3.2 GB.

Read the source: huggingface.co ↗In the 15 November issue →