Adversarial GLUE: A Multi-Task Benchmark for Robustness Evaluation of Language Models.
Boxin WangChejian XuShuohang WangZhe GanYu ChengJianfeng GaoAhmed Hassan AwadallahBo LiPublished in: CoRR (2021)
Keyphrases
- language model
- multi task
- language modeling
- multi task learning
- probabilistic model
- n gram
- learning tasks
- test collection
- information retrieval
- language modelling
- query expansion
- retrieval model
- document retrieval
- multiple tasks
- statistical language models
- document ranking
- smoothing methods
- feature selection
- gaussian processes
- transfer learning
- sparse learning
- learning problems
- relevance model
- active learning
- language models for information retrieval
- document collections
- reinforcement learning
- decision trees