Massive Text Embedding Benchmark: A global effort to expand text embedding evaluation to all languages

Enevoldsen, Kenneth Christian (PI)
Bernstorff, Martin (Deltager)
Kardos, Márton (Deltager)
Kerboua, Imene (Projektkoordinator)
Schaeffer, Marion (Deltager)
Xiao, Shitao (Deltager)
Cassano, Federico (Deltager)
Li, Wen-Ding (Deltager)
Rystrøm, Jonathan (Deltager)
Lee, Taemin (Deltager)
Zhang, Xin (Deltager)
Weller, Orion (Deltager)

Institut for Kultur og Samfund - Center for Humanities Computing Aarhus
Institut for Klinisk Medicin - Afdeling for Depression og Angst
Institut for Folkesundhed - Epidemiologi

Projekter: Projekt › Forskning

Beskrivelse

Massive Text Embedding Benchmark (MMTEB) is a global effort to expand text embedding evaluation to all languages with more than 50 contributors. The benchmark seeks to evaluate the quality of embeddings of text, e.g. used for search, retrieval etc.

Kort titel	Massive Text Embedding Benchmark
Akronym	MMTEB
Status	Igangværende
Effektiv start/slut dato	01/04/2024 → …

Adgang til projekt

https://github.com/embeddings-benchmark/mteb/blob/main/docs/mmteb/readme.md

Massive Text Embedding Benchmark: A global effort to expand text embedding evaluation to all languages

Projektdetaljer

Beskrivelse

Adgang til projekt

Fingerprint