A probabilistic polarity scoring method for Latent Semantic Scaling uses masked language models to compute seed-word occurrence probabilities, claimed to outperform spatial models in accuracy, interpretability, and consistency.
These seed words are not optimal but allow evaluation of models in situations where seed words chosen by users are relevant but not necessarily most extreme words.10
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
A New Semisupervised Technique for Polarity Analysis using Masked Language Models
A probabilistic polarity scoring method for Latent Semantic Scaling uses masked language models to compute seed-word occurrence probabilities, claimed to outperform spatial models in accuracy, interpretability, and consistency.