Supervised Learning of Universal Sentence Representations from Natural Language Inference Data

Benchmark Model Rank Results
natural-language-inference-on-snli4096D BiLSTM with max-pooling#54% Test Accuracy: 84.5% Train Accuracy: 85.6Parameters: 40m
semantic-textual-similarity-on-mrpcInferSent#30Accuracy: 76.2%F1: 83.1%