PaLM 540B (finetuned)

Reported on 6 benchmarks across 3 tasks · 1 paper · 4 SOTA

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

Natural Language Processing6 results

Question AnsweringonCOPA
Accuracy· 2022-04-05
100
SOTA
PaLM: Scaling Language Modeling with Pathways arXiv:2204.02311
Question AnsweringonMultiRC
EM· 2022-04-05
69.2
SOTA
PaLM: Scaling Language Modeling with Pathways arXiv:2204.02311
Question AnsweringonMultiRC
F1· 2022-04-05
90.1
SOTA
PaLM: Scaling Language Modeling with Pathways arXiv:2204.02311
Common Sense ReasoningonReCoRD
F1· 2022-04-05
94.6
best: 96.4 (Turing NLR v5 XXL 5.4B (fine-tuned))
SOTA
PaLM: Scaling Language Modeling with Pathways arXiv:2204.02311
Common Sense ReasoningonReCoRD
EM· 2022-04-05
94
best: 95.9 (Turing NLR v5 XXL 5.4B (fine-tuned))
PaLM: Scaling Language Modeling with Pathways arXiv:2204.02311
Word Sense DisambiguationonWords in Context
Accuracy· 2022-04-05
78.8
best: 85.3 (COSINE + Transductive Learning)
PaLM: Scaling Language Modeling with Pathways arXiv:2204.02311