Tasks SotA Datasets Papers Methods Submit About

Papers With Code 2

A community resource for machine learning research: papers, code, benchmarks, and state-of-the-art results.

Explore

Notable Benchmarks All SotA Datasets Papers Methods

Community

Submit Results About

Data sourced from the PWC Archive (CC-BY-SA 4.0). Built by the community, for the community.

Models/ST-MoE-L 4.1B (fine-tuned)

ST-MoE-L 4.1B (fine-tuned)

Reported on 10 benchmarks across 5 tasks · 1 paper · 1 SOTA

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

Natural Language Processing10 results

Natural Language InferenceonCommitmentBank
Accuracy· 2022-02-17
98.2
best: 100 (PaLM 540B (finetuned))
SOTA
ST-MoE: Designing Stable and Transferable Sparse Expert Models arXiv:2202.08906
Question AnsweringonCOPA
Accuracy· 2022-02-17
91
best: 100 (PaLM 540B (finetuned) )
ST-MoE: Designing Stable and Transferable Sparse Expert Models arXiv:2202.08906
Question AnsweringonMultiRC
F1· 2022-02-17
86
best: 90.1 (PaLM 540B (finetuned) )
ST-MoE: Designing Stable and Transferable Sparse Expert Models arXiv:2202.08906
Question AnsweringonBoolQ
Accuracy· 2022-02-17
88.6
best: 99.87 (Mistral-Nemo 12B (HPT))
ST-MoE: Designing Stable and Transferable Sparse Expert Models arXiv:2202.08906
Common Sense ReasoningonWinoGrande
Accuracy· 2022-02-17
81.7
best: 96.1 (ST-MoE-32B 269B (fine-tuned))
ST-MoE: Designing Stable and Transferable Sparse Expert Models arXiv:2202.08906
Common Sense ReasoningonARC (Challenge)
Accuracy· 2022-02-17
56.9
best: 96.4 (GPT-4 (few-shot, k=25))
ST-MoE: Designing Stable and Transferable Sparse Expert Models arXiv:2202.08906
Common Sense ReasoningonARC (Easy)
Accuracy· 2022-02-17
75.4
best: 95.2 (ST-MoE-32B 269B (fine-tuned))
ST-MoE: Designing Stable and Transferable Sparse Expert Models arXiv:2202.08906
Common Sense ReasoningonReCoRD
EM· 2022-02-17
88.9
best: 95.9 (Turing NLR v5 XXL 5.4B (fine-tuned))
ST-MoE: Designing Stable and Transferable Sparse Expert Models arXiv:2202.08906
Word Sense DisambiguationonWords in Context
Accuracy· 2022-02-17
74
best: 85.3 (COSINE + Transductive Learning)
ST-MoE: Designing Stable and Transferable Sparse Expert Models arXiv:2202.08906
Coreference ResolutiononWinograd Schema Challenge
Accuracy· 2022-02-17
93.3
best: 100 (PaLM 540B (fine-tuned))
ST-MoE: Designing Stable and Transferable Sparse Expert Models arXiv:2202.08906