Acquiring Clean Language Models from Backdoor Poisoned Datasets by Downscaling Frequency Space.
Zongru WuZhuosheng ZhangPengzhou ChengGongshen LiuPublished in: CoRR (2024)
Keyphrases
- language model
- language modeling
- n gram
- speech recognition
- probabilistic model
- document retrieval
- information retrieval
- test collection
- statistical language models
- retrieval model
- context sensitive
- language models for information retrieval
- language modelling
- query expansion
- document ranking
- smoothing methods
- query terms
- vector space
- ad hoc information retrieval
- translation model
- search engine