BERT Large

Reported on 3 benchmarks across 3 tasks · 2 papers · 2 SOTA

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

Natural Language Processing3 results

Question AnsweringonCODAH
Accuracy· uses extra data· 2019-04-08
69.6
best: 84 (G-DAUG-Combo + RoBERTa-Large)
SOTA
CODAH: An Adversarially Authored Question-Answer Dataset for Common Sense arXiv:1904.04365
Common Sense ReasoningonCODAH
Accuracy· uses extra data· 2019-04-08
69.6
SOTA
CODAH: An Adversarially Authored Question-Answer Dataset for Common Sense arXiv:1904.04365
Sentiment AnalysisonSST-5 Fine-grained classification
Accuracy· 2019-10-04
55.5
best: 62.27 (Llama-3.3-70B + CAPO)
Fine-grained Sentiment Classification using BERT arXiv:1910.03474