Proto-lm: A Prototypical Network-Based Framework for Built-in Interpretability in Large Language Models.
Sean XieSoroush VosoughiSaeed HassanpourPublished in: EMNLP (Findings) (2023)
Keyphrases
- language model
- language modeling
- probabilistic model
- query expansion
- document retrieval
- information retrieval
- n gram
- retrieval model
- relevance model
- context sensitive
- language modelling
- speech recognition
- statistical language models
- ad hoc information retrieval
- pseudo relevance feedback
- prediction accuracy
- text retrieval
- smoothing methods
- statistical language modeling
- language models for information retrieval
- okapi bm
- visual features
- hidden markov models
- decision trees