Token-level editing of human text with a high-confidence threshold yields modest performance gains over the original data and avoids the sharp degradation seen with purely synthetic data.
Picor: Multi-task deep reinforcement learning with policy correction
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
2024 1verdicts
REJECT 1roles
background 1polarities
unclear 1representative citing papers
citing papers explorer
-
How to Synthesize Text Data without Model Collapse?
Token-level editing of human text with a high-confidence threshold yields modest performance gains over the original data and avoids the sharp degradation seen with purely synthetic data.