A contrastive fine-tuning method that adds frozen semantic-embedding negatives reduces topic leakage in authorship embeddings and improves accuracy on topic-heavy attribution benchmarks.
Cross-genre author profile prediction using stylometry-based approach
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2024 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Isolating authorship from content with semantic embeddings and contrastive learning
A contrastive fine-tuning method that adds frozen semantic-embedding negatives reduces topic leakage in authorship embeddings and improves accuracy on topic-heavy attribution benchmarks.