opensearch-project/opensearch-neural-sparse-encoding-doc-v2-distill
The model should be selected considering search relevance, model inference and retrieval efficiency(FLOPS). We benchmark models' zero-shot performance on a subset of BEIR benchmark: TrecCovid,NFCorpus,NQ,HotpotQA,FiQA,ArguAna,Touche,DBPedia,SCIDOCS,FEVER,Climate FEVER,SciFact,Quora.
Benchmarks
CQADupstackPhysicsRetrieval
Duplicate question retrieval from StackExchange Physics
Corpus: 38,314 Queries: 1,039
Performance L4 b1 c16
Corpus TPS 31.0K
Corpus p50 60.2ms
Query TPS 4.1K
Query p50 41.5ms
CosQA
Code search with natural language queries
Corpus: 6,267 Queries: 500
Performance L4 b1 c16
Corpus TPS 17.8K
Corpus p50 46.4ms
Query TPS 2.6K
Query p50 37.3ms
FiQA2018
Financial opinion mining and question answering
Corpus: 57,599 Queries: 648
Performance L4 b1 c16
Corpus TPS 43.5K
Corpus p50 56.2ms
Query TPS 4.2K
Query p50 42.8ms
LegalBenchConsumerContractsQA
Question answering on consumer contracts
Corpus: 153 Queries: 396
Performance L4 b1 c16
Corpus TPS 104.9K
Corpus p50 72.6ms
Query TPS 6.9K
Query p50 37.3ms
NFCorpus
Biomedical literature search from NutritionFacts.org
Corpus: 3,593 Queries: 323
Quality
ndcg at 10 0.3398
map at 10 0.1303
mrr at 10 0.5489
Performance L4 b1 c16
Corpus TPS 56.8K
Corpus p50 70.8ms
Query TPS 1.8K
Query p50 41.7ms
SCIDOCS
Citation prediction, document classification, and recommendation for scientific papers
Corpus: 25,656 Queries: 1,000
Performance L4 b1 c16
Corpus TPS 48.7K
Corpus p50 58.5ms
Query TPS 4.7K
Query p50 37.7ms
SciFact
Scientific claim verification using research literature
Corpus: 5,183 Queries: 300
Performance L4 b1 c16
Corpus TPS 60.2K
Corpus p50 66.4ms
Query TPS 5.8K
Query p50 44.1ms
StackOverflowQA
Programming question answering from Stack Overflow
Corpus: 19,931 Queries: 1,994
Performance L4 b1 c16
Corpus TPS 49.6K
Corpus p50 70.8ms
Query TPS 77.6K
Query p50 52.3ms