Adversarial Robustness On Advglue

Métriques

Accuracy

Résultats

Résultats de performance de divers modèles sur ce benchmark

		Paper Title
DeBERTa (single model)	0.6086	Adversarial GLUE: A Multi-Task Benchmark for Robustness Evaluation of Language Models
ALBERT (single model)	0.5922	Adversarial GLUE: A Multi-Task Benchmark for Robustness Evaluation of Language Models
T5 (single model)	0.5682	Adversarial GLUE: A Multi-Task Benchmark for Robustness Evaluation of Language Models
SMART_RoBERTa (single model)	0.5371	Adversarial GLUE: A Multi-Task Benchmark for Robustness Evaluation of Language Models
FreeLB (single model)	0.5048	Adversarial GLUE: A Multi-Task Benchmark for Robustness Evaluation of Language Models
RoBERTa (single model)	0.5021	Adversarial GLUE: A Multi-Task Benchmark for Robustness Evaluation of Language Models
InfoBERT (single model)	0.4603	Adversarial GLUE: A Multi-Task Benchmark for Robustness Evaluation of Language Models
ELECTRA (single model)	0.4169	Adversarial GLUE: A Multi-Task Benchmark for Robustness Evaluation of Language Models
BERT (single model)	0.3369	Adversarial GLUE: A Multi-Task Benchmark for Robustness Evaluation of Language Models
SMART_BERT (single model)	0.3029	Adversarial GLUE: A Multi-Task Benchmark for Robustness Evaluation of Language Models

0 of 10 row(s) selected.

Adversarial Robustness On Advglue | SOTA | HyperAI