Natural Language Inference On Multinli
评估指标
Matched
Mismatched
评测结果
各个模型在此基准测试上的表现结果
比较表格
| 模型名称 | Matched | Mismatched |
|---|---|---|
| ernie-20-a-continual-pre-training-framework | 88.7 | 88.8 |
| exploring-the-limits-of-transfer-learning | 87.1 | 86.2 |
| not-all-layers-are-equally-as-important-every | 84.4 | 84.5 |
| training-complex-models-with-multi-task-weak | 87.6 | 87.2 |
| exploring-the-limits-of-transfer-learning | 91.4 | 91.2 |
| learning-general-purpose-distributed-sentence | 71.4 | 71.3 |
| roberta-a-robustly-optimized-bert-pretraining | 90.8 | - |
| charformer-fast-character-transformers-via | 83.7 | 84.4 |
| generative-pretrained-structured-transformers | 81.8 | 82.0 |
| lm-cppf-paraphrasing-guided-data-augmentation | - | - |
| exploring-the-limits-of-transfer-learning | - | 91.7 |
| smart-robust-and-efficient-fine-tuning-for | - | - |
| exploring-the-limits-of-transfer-learning | 92.0 | - |
| informer-transformer-likes-informed-attention | 86.28 | 86.34 |
| 190910351 | 84.6 | 83.2 |
| glue-a-multi-task-benchmark-and-analysis | 72.2 | 72.1 |
| combining-similarity-features-and-deep | 70.7 | 71.1 |
| structbert-incorporating-language-structures | 91.1 | 90.7 |
| lamini-lm-a-diverse-herd-of-distilled-models | 36.5 | 37 |
| first-train-to-generate-then-generate-to | 89.8 | - |
| q8bert-quantized-8bit-bert | 85.6 | - |
| pay-attention-to-mlps | 86.2 | 86.5 |
| baseline-needs-more-love-on-simple-word | 68.2 | 67.7 |
| adversarial-self-attention-for-language | 85 | - |
| lamini-lm-a-diverse-herd-of-distilled-models | 72.4 | 72 |
| smart-robust-and-efficient-fine-tuning-for | 92.0 | 91.7 |
| exploring-the-limits-of-transfer-learning | - | 89.6 |
| ernie-20-a-continual-pre-training-framework | 86.1 | 85.5 |
| deberta-decoding-enhanced-bert-with | 91.1 | 91.1 |
| not-all-layers-are-equally-as-important-every | 79.2 | 79.9 |
| smart-robust-and-efficient-fine-tuning-for | - | - |
| smart-robust-and-efficient-fine-tuning-for | - | - |
| spanbert-improving-pre-training-by | 88.1 | - |
| fnet-mixing-tokens-with-fourier-transforms | 78 | 76 |
| llm-int8-8-bit-matrix-multiplication-for | 90.2 | - |
| smart-robust-and-efficient-fine-tuning-for | - | - |
| exploring-the-limits-of-transfer-learning | 82.4 | 82.3 |
| attention-boosted-sequential-inference-model | 73.9 | 73.9 |
| ernie-enhanced-language-representation-with | 84.0 | 83.2 |
| lamini-lm-a-diverse-herd-of-distilled-models | 61.4 | 61 |
| xlnet-generalized-autoregressive-pretraining | 90.8 | - |
| lamini-lm-a-diverse-herd-of-distilled-models | 67.5 | 69.3 |
| first-train-to-generate-then-generate-to | 92.6 | - |
| q-bert-hessian-based-ultra-low-precision | 87.8 | - |
| combining-similarity-features-and-deep | 70.7 | 70.5 |
| squeezebert-what-can-computer-vision-teach | 82.0 | 81.1 |
| roberta-a-robustly-optimized-bert-pretraining | - | 90.2 |
| smart-robust-and-efficient-fine-tuning-for | - | - |
| improving-language-understanding-by | 82.1 | 81.4 |
| combining-similarity-features-and-deep | 71.4 | 72.2 |
| big-bird-transformers-for-longer-sequences | 87.5 | - |
| lamini-lm-a-diverse-herd-of-distilled-models | 54.7 | 55.8 |
| what-do-questions-exactly-ask-mfae-duplicate | 82.31 | 81.43 |
| 模型 54 | 92.6 | 92.4 |
| adversarial-self-attention-for-language | 88 | - |
| improving-multi-task-deep-neural-networks-via | 87.9 | 87.4 |
| 模型 57 | 82.1 | 81.4 |
| fnet-mixing-tokens-with-fourier-transforms | 88 | 88 |
| bert-pre-training-of-deep-bidirectional | 86.7 | 85.9 |
| not-all-layers-are-equally-as-important-every | 78 | 78.8 |
| albert-a-lite-bert-for-self-supervised | 91.3 | - |
| how-to-train-bert-with-an-academic-budget | 84.4 | 83.8 |
| multi-task-deep-neural-networks-for-natural | 86.7 | 86.0 |
| exploring-the-limits-of-transfer-learning | 89.9 | - |
| not-all-layers-are-equally-as-important-every | 83 | 83.4 |
| a-statistical-framework-for-low-bitwidth | 89.9 | - |
| 190910351 | 82.5 | 81.8 |