Fine-tuned T5-small and BART can lower the accuracy of GPT-text classifiers, but retraining the classifiers on the transformed text restores detection accuracy.
XGBoost: A scalable tree boosting sys- tem
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.CL 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
unclear 1representative citing papers
citing papers explorer
-
Transforming Chatbot Text: A Sequence-to-Sequence Approach
Fine-tuned T5-small and BART can lower the accuracy of GPT-text classifiers, but retraining the classifiers on the transformed text restores detection accuracy.