Team dina at SemEval-2022 Task 8: Pre-trained Language Models as Baselines for Semantic Similarity.
Dina PisarevskayaArkaitz ZubiagaPublished in: SemEval@NAACL (2022)
Keyphrases
- semantic similarity
- language model
- pre trained
- word sense disambiguation
- language modeling
- wordnet
- co occurrence
- training data
- vector space model
- n gram
- information retrieval
- semantic relatedness
- training examples
- retrieval model
- similarity measure
- speech recognition
- probabilistic model
- document retrieval
- query expansion
- semantic information
- test collection
- pseudo relevance feedback
- control signals
- cross lingual
- query terms
- decision trees
- natural language processing
- supervised learning
- active learning
- xml documents
- tf idf
- word sense
- relevance model
- high level