Towards Tracing Knowledge in Language Models Back to the Training Data.
Ekin AkyürekTolga BolukbasiFrederick LiuBinbin XiongIan TenneyJacob AndreasKelvin GuuPublished in: EMNLP (Findings) (2022)
Keyphrases
- language model
- training data
- language modeling
- prior knowledge
- probabilistic model
- n gram
- document retrieval
- domain knowledge
- speech recognition
- information retrieval
- statistical language models
- retrieval model
- language modelling
- learning algorithm
- smoothing methods
- test collection
- context sensitive
- knowledge discovery
- ad hoc information retrieval
- language model for information retrieval
- machine learning
- pseudo relevance feedback
- translation model
- document ranking
- training set
- decision trees