NMSLIB author publicly slams new vector search preprint for downplaying prior work
srchvrs · x · 2026-11-05
NMSLIB author srchvrs calls a new vector-search preprint's treatment "extremely unfair": NMSLIB was the first widely used ANN library supporting dense and sparse representations, multiple similarity measures, and graph-based indexing for both, with its 2016 CIKM paper showing graph-based sparse retrieval competitiveness long before the 2023 BigANN challenge. He says he privately flagged this in January and got acknowledgment, yet the preprint never names NMSLIB as a library and cites their paper to back an unsubstantiated claim that multi-representation support "reduces performance on each." The authors respond the sentence's subject was hnswlib and that their dense/sparse/multivector scope only warrants citing nmslib in the intro.
Related event: NMSLIB Author Protests New Vector Search Paper Downplaying His Work(4 posts)→
More from coding & agent
- Max out Claude thinking and ultracode on every prompt, says dev — curious_vii · 2026-11-05
- After 100+ team interviews, Swytchcode argues MCP is for building agents, not running them — curiousblack99 · 2026-11-05
- Yuchen declares 'Codex won the agent war' — Yuchenj_UW · 2026-11-05
- Interactive home-emergency trainer built with Opus 5.5, Blender and Three.js — techartist_ · 2026-11-05
- Developer proposes a 12-level AI coding autonomy scale based on who does the work — nimareq · 2026-11-05
- Claude Cookbook shows how to build a nightly "dreaming" memory service for agents — brada · 2026-11-05