First stance detection study on prediction market commentary finds market context raises 3-class Anti recall from 0.10 to 0.45 while 50% counterfactual augmentation is optimal and full augmentation hurts performance.
Title resolution pending
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
-
Stance Detection in Prediction Markets: Addressing Imbalanced Trader Commentary via Counterfactual Augmentation and Market Context
First stance detection study on prediction market commentary finds market context raises 3-class Anti recall from 0.10 to 0.45 while 50% counterfactual augmentation is optimal and full augmentation hurts performance.