BERTlarge-flow (target)

Reported on 6 benchmarks across 1 task · 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 Processing6 results

Semantic Textual SimilarityonSTS16
Spearman Correlation· 2020-11-02
0.7763
best: 0.87 (AnglE-LLaMA-7B-v2)
SOTA
On the Sentence Embeddings from Pre-trained Language Models arXiv:2011.05864
Semantic Textual SimilarityonSTS14
Spearman Correlation· 2020-11-02
0.6942
best: 0.8689 (AnglE-LLaMA-13B)
On the Sentence Embeddings from Pre-trained Language Models arXiv:2011.05864
Semantic Textual SimilarityonSTS15
Spearman Correlation· 2020-11-02
0.7492
best: 0.9004 (PromptEOL+CSE+LLaMA-30B)
On the Sentence Embeddings from Pre-trained Language Models arXiv:2011.05864
Semantic Textual SimilarityonSTS13
Spearman Correlation· 2020-11-02
0.7339
best: 0.9058 (AnglE-LLaMA-7B)
On the Sentence Embeddings from Pre-trained Language Models arXiv:2011.05864
Semantic Textual SimilarityonSTS Benchmark
Spearman Correlation· 2020-11-02
0.7226
best: 0.931 (Mnet-Sim)
On the Sentence Embeddings from Pre-trained Language Models arXiv:2011.05864
Semantic Textual SimilarityonSTS12
Spearman Correlation· 2020-11-02
0.652
best: 0.802 (PromptEOL+CSE+OPT-13B)
On the Sentence Embeddings from Pre-trained Language Models arXiv:2011.05864