Enhancing Interpretability Using Human Similarity Judgements to Prune Word Embeddings.
Natalia Flechas ManriqueWanqian BaoAurélie HerbelotUri HassonPublished in: BlackboxNLP@EMNLP (2023)
Keyphrases
- word similarity
- sentence similarity
- n gram
- similarity measure
- human subjects
- semantic similarity
- co occurrence
- search space
- information extraction
- similarity measurement
- distance measure
- prediction accuracy
- human behavior
- neural network
- high dimensional data
- euclidean distance
- vector space
- similarity metric
- binary codes
- word segmentation
- word pairs
- pairwise