Almanac
technique

counterfactual data augmentation

techniqueactiveprovisionalcounterfactual-data-augmentation-d22a9ba6·1 events·first seen 20d ago

Aliases: counterfactual data augmentation

Co-occurring entities

More like this (12)

Recent events (1)

4arXiv · cs.CL·20d ago·source ↗

Stance Detection in Prediction Market Commentary via Counterfactual Augmentation and Market Context

This paper introduces the first stance detection system applied to prediction market commentary (Polymarket), addressing extreme class imbalance (8.7% anti-market comments) through LLM-driven counterfactual augmentation using the Anthropic API. RoBERTa-base is fine-tuned across a 4×3 ablation covering input configurations and augmentation doses. Key findings: market context is the dominant factor (raising 3-class Anti recall from 0.10 to 0.45), 50% synthetic augmentation is optimal, and full augmentation (100%) consistently degrades performance. Attention-based interpretability supports all three findings mechanistically.