AWhat changed
Unified evaluation across embedding task types and extended it through community-reviewed multilingual tasks, caching and document/task downsampling to reduce repeated evaluation work.
Scientific scope
Unified evaluation across embedding task types and extended it through community-reviewed multilingual tasks, caching and document/task downsampling to reduce repeated evaluation work. MTEB v1 and MMTEB v1 have different task inventories and evaluation pipelines; original MTEB classification uses the stated kNN protocol. MMTEB reduced sets preserve rankings empirically for studied models, not universally for future models. Managing-team and source authorship points are not WRS effort/scoring. Incidental medical-domain texts do not constitute a medical achievement.
Evidence
The linked original sources describe the stated contribution and its evidence limits. WRS has not independently rerun the experiments or certified a proof. MTEB v1 and MMTEB v1 have different task inventories and evaluation pipelines; original MTEB classification uses the stated kNN protocol. MMTEB reduced sets preserve rankings empirically for studied models, not universally for future models. Managing-team and source authorship points are not WRS effort/scoring. Incidental medical-domain texts do not constitute a medical achievement.
Contributors
- Source-listed authors and technical contributors · SCIENCE: Source-listed authors or explicitly credited technical contributors for the cited versions. Family entries retain source-specific attribution; a combined list is not one original-paper byline or equal individual work. Names alone do not establish resolved identities.
Provisional WRS assessment, edition 2026-10-16.1. Independent expert calibration is pending. This grade does not certify correctness or assign ownership.
Full scientific record · Browse contribution grades · Request a correction