Fine-tuning optical structure recognizers with labeled real depictions, even at 5 to 10 percent of the training mixture, is the most effective intervention for closing the synthetic-to-real gap.
Journal of Chemical Information and Modeling , volume=
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Real Data Closes Synthetic-to-Real Gap in Optical Chemical Structure Recognition
Fine-tuning optical structure recognizers with labeled real depictions, even at 5 to 10 percent of the training mixture, is the most effective intervention for closing the synthetic-to-real gap.