Natural Language Inference On Snli
المقاييس
% Test Accuracy
% Train Accuracy
Parameters
النتائج
نتائج أداء النماذج المختلفة على هذا المعيار القياسي
جدول المقارنة
| اسم النموذج | % Test Accuracy | % Train Accuracy | Parameters |
|---|---|---|---|
| self-explaining-structures-improve-nlp-models | 92.3 | ? | 355m+ |
| distance-based-self-attention-network-for | 86.3 | 89.6 | 4.7m |
| combining-similarity-features-and-deep | 84.4 | - | - |
| learning-to-compose-task-specific-tree | 85.6 | 91.2 | 2.9m |
| multi-task-deep-neural-networks-for-natural | 91.6 | 97.2 | 330m |
| parameter-re-initialization-through-cyclical | 86.73 | - | - |
| i-know-what-you-want-semantic-learning-for | 91.3 | 95.7 | 308m |
| conditionally-adaptive-multi-task-learning | 92.1 | 92.6 | 340m |
| order-embeddings-of-images-and-language | 81.4 | 98.8 | 15m |
| a-decomposable-attention-model-for-natural | 86.8 | 90.5 | 580k |
| smart-robust-and-efficient-fine-tuning-for | - | - | - |
| learning-natural-language-inference-using | 85.0 | 85.9 | 2.8m |
| enhancing-sentence-embedding-with-generalized | 86.6 | 94.9 | 65m |
| enhanced-lstm-for-natural-language-inference | 88.0 | - | - |
| reinforced-self-attention-network-a-hybrid-of | 86.3 | 92.6 | 3.1m |
| deep-fusion-lstms-for-text-semantic-matching | 84.6 | 85.2 | 320k |
| recurrent-neural-network-based-sentence | 85.5 | 90.5 | 12m |
| dr-bilstm-dependent-reading-bidirectional | 88.5 | 94.1 | 7.5m |
| shortcut-stacked-sentence-encoders-for-multi | 85.7 | 89.8 | 9.7m |
| deep-contextualized-word-representations | 89.3 | 92.1 | 40m |
| reading-and-thinking-re-read-lstm-unit-for | 87.5 | 90.7 | 2.0m |
| semantics-aware-bert-for-language | 91.9 | 94.4 | 339m |
| what-do-questions-exactly-ask-mfae-duplicate | 90.07 | 93.18 | - |
| a-fast-unified-model-for-parsing-and-sentence | 80.6 | 83.9 | 3.0m |
| natural-language-inference-over-interaction-1 | 88.9 | 92.3 | 17m |
| i-know-what-you-want-semantic-learning-for | 89.1 | 89.1 | 6.1m |
| a-large-annotated-corpus-for-learning-natural | 50.4 | 49.4 | |
| smart-robust-and-efficient-fine-tuning-for | 91.7 | - | - |
| smart-robust-and-efficient-fine-tuning-for | - | - | - |
| compare-compress-and-propagate-enhancing | 85.9 | 87.3 | 3.7m |
| baseline-needs-more-love-on-simple-word | 83.8 | - | - |
| splitee-early-exit-in-deep-neural-networks | - | - | - |
| semantic-sentence-matching-with-densely | 90.1 | 95.0 | 53.3m |
| a-fast-unified-model-for-parsing-and-sentence | 83.2 | 89.2 | 3.7m |
| a-decomposable-attention-model-for-natural | 86.8 | 90.5 | 580k |
| neural-tree-indexers-for-text-understanding | 87.3 | 88.5 | 3.2m |
| learning-to-compose-task-specific-tree | 86.0 | 93.1 | 10m |
| entailment-as-few-shot-learner | 93.1 | - | 355 |
| compare-compress-and-propagate-enhancing | 88.5 | 89.8 | 4.7m |
| dr-bilstm-dependent-reading-bidirectional | 89.3 | 94.8 | 45m |
| self-explaining-structures-improve-nlp-models | 92.3 | - | 340 |
| long-short-term-memory-networks-for-machine | 85.7 | 87.3 | 1.7m |
| bilateral-multi-perspective-matching-for | 87.5 | 90.9 | 1.6m |
| النموذج 44 | 89.6 | 96.1 | 79m |
| combining-similarity-features-and-deep | 84.5 | - | - |
| simple-and-effective-text-matching-with-1 | 88.9 | 94.0 | 2.8m |
| stochastic-answer-networks-for-natural | 88.5 | 93.3 | 3.5m |
| learning-natural-language-inference-with-lstm | 86.1 | 92.0 | 1.9m |
| natural-language-inference-by-tree-based | 82.1 | 83.3 | 3.5m |
| first-train-to-generate-then-generate-to | 93.5 | - | - |
| learned-in-translation-contextualized-word | 88.1 | 88.5 | 22m |
| supervised-learning-of-universal-sentence | 84.5 | 85.6 | 40m |
| النموذج 53 | 87.5 | 90.7 | 2.0m |
| star-transformer | 86.0 | - | - |
| neural-semantic-encoders | 84.6 | 86.2 | 3.0m |
| multiway-attention-networks-for-modeling | 89.4 | 95.5 | 58m |
| smart-robust-and-efficient-fine-tuning-for | - | - | - |
| disan-directional-self-attention-network-for | 85.6 | 91.1 | 2.4m |
| النموذج 59 | 88.8 | 95.4 | 9.2m |
| semantic-sentence-matching-with-densely | 88.9 | 93.1 | 6.7m |
| neural-natural-language-inference-models | 89.1 | 93.6 | 43m |
| reasoning-about-entailment-with-neural | 83.5 | 85.3 | 250k |
| modelling-interaction-of-sentence-pair-with | 85.1 | 86.7 | 190k |
| neural-semantic-encoders | 85.4 | 86.9 | 3.2m |
| dynamic-self-attention-computing-attention | 87.4 | 89.0 | 7.0m |
| smart-robust-and-efficient-fine-tuning-for | - | - | - |
| learning-natural-language-inference-using | 84.2 | 84.5 | 2.8m |
| neural-natural-language-inference-models | 88.6 | 94.1 | 4.3m |
| delta-a-deep-learning-based-language | 80.7 | - | - |
| a-large-annotated-corpus-for-learning-natural | 77.6 | 84.8 | 220k |
| discourse-marker-augmented-network-with-1 | 88.8 | 95.4 | 9.2m |
| entailment-as-few-shot-learner | 93.1 | ? | 355m |
| bilateral-multi-perspective-matching-for | 88.8 | 93.2 | 6.4m |
| deim-an-effective-deep-encoding-and | 88.9 | 92.6 | 22m |
| neural-tree-indexers-for-text-understanding | 83.4 | 82.5 | 4.0m |
| long-short-term-memory-networks-for-machine | 86.3 | 88.5 | 3.4m |
| natural-language-inference-with-hierarchical | 86.6 | 89.9 | 22m |
| a-large-annotated-corpus-for-learning-natural | 78.2 | 99.7 | |
| deep-contextualized-word-representations | 88.7 | 91.6 | 8.0m |
| compare-compress-and-propagate-enhancing | 89.3 | 92.5 | 17.5m |
| learning-natural-language-inference-using | 83.3 | 86.4 | 2.0m |
| discourse-marker-augmented-network-with-1 | 89.6 | 96.1 | 79m |
| first-train-to-generate-then-generate-to | 94.7 | - | - |
| multiway-attention-networks-for-modeling | 88.3 | 94.5 | 14m |
| a-decomposable-attention-model-for-natural | 86.3 | 89.5 | 380k |
| enhanced-lstm-for-natural-language-inference | 88.6 | 93.5 | 7.7m |
| improving-language-understanding-by | 89.9 | 96.6 | 85m |
| dynamic-self-attention-computing-attention | 86.8 | 87.3 | 2.1m |
| a-decomposable-attention-model-for-natural | 86.3 | 89.5 | 380k |
| dynamic-meta-embeddings-for-improved-sentence | 86.7 | 91.6 | 9m |
| shortcut-stacked-sentence-encoders-for-multi | 86.0 | 91.0 | 29m |
| combining-similarity-features-and-deep | 84.8 | - | - |
| multi-task-deep-neural-networks-for-natural | 90.5 | 99.1 | 220 |
| cell-aware-stacked-lstms-for-modeling | 87 | - | - |
| natural-language-inference-over-interaction-1 | 88.0 | 91.2 | 4.4m |
| semantic-sentence-matching-with-densely | 86.5 | 91.4 | 5.6m |
| النموذج 97 | 85.9 | – | – |
| attention-boosted-sequential-inference-model | 88.1 | - | - |