AST-Probe: Recovering abstract syntax trees from hidden representations of pre-trained language models.
José Antonio Hernández LópezMartin WeyssowJesús Sánchez CuadradoHouari A. SahraouiPublished in: CoRR (2022)
Keyphrases
- language model
- pre trained
- language modeling
- word clouds
- document retrieval
- probabilistic model
- training data
- information retrieval
- n gram
- test collection
- query expansion
- speech recognition
- language modelling
- training examples
- retrieval model
- smoothing methods
- statistical language models
- control signals
- relevance model
- supervised learning
- small number
- document ranking
- data sets
- text mining
- learning algorithm
- image classification