RoAST: Robustifying Language Models via Adversarial Perturbation with Selective Training.
Jaehyung KimYuning MaoRui HouHanchao YuDavis LiangPascale FungQifan WangFuli FengLifu HuangMadian KhabsaPublished in: CoRR (2023)
Keyphrases
- language model
- language modeling
- n gram
- document retrieval
- language modelling
- probabilistic model
- information retrieval
- speech recognition
- statistical language models
- query expansion
- retrieval model
- document ranking
- context sensitive
- test collection
- language models for information retrieval
- ad hoc information retrieval
- word error rate
- okapi bm
- smoothing methods
- vector space model
- language model for information retrieval
- term dependencies
- relevance model
- document length
- translation model
- training set