SPARTA

Reported on 5 benchmarks across 5 tasks · 3 papers · 1 SOTA

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Natural Language Processing5 results

Open-Domain Question AnsweringonSQuAD1.1 dev
EM· 2020-09-28
59.3
SOTA
SPARTA: Efficient Open-Domain Question Answering via Sparse Transformer Matching Retrieval arXiv:2009.13013
Cross-LingualonSV-Ident
mAP@10· 2022-09-19
11.27
best: 18.93 (sentence-transformers/distiluse-base-multilingual-cased-v1)
Overview of the SV-Ident 2022 Shared Task on Survey Variable Identification in Social Science Publications arXiv:2209.09062
Cross-Lingual Entity LinkingonSV-Ident
mAP@10· 2022-09-19
11.27
best: 18.93 (sentence-transformers/distiluse-base-multilingual-cased-v1)
Overview of the SV-Ident 2022 Shared Task on Survey Variable Identification in Social Science Publications arXiv:2209.09062
Information RetrievalonMSMARCO (BEIR)
nDCG@10· 2021-04-17
0.351
best: 0.413 (BM25+CE)
BEIR: A Heterogenous Benchmark for Zero-shot Evaluation of Information Retrieval Models arXiv:2104.08663
Question AnsweringonSQuAD1.1 dev
EM· 2020-09-28
59.3
best: 90.06 (T5-11B)
SPARTA: Efficient Open-Domain Question Answering via Sparse Transformer Matching Retrieval arXiv:2009.13013