Entity · technique

stance detection

techniqueactivestance-detection-0e2d5453·1 events·first seen May 28, 2026

Aliases: stance detection

Co-occurring entities

RoBERTa counterfactual data augmentation Polymarket Anthropic

More like this (12)

Safety Detection Classifier hate speech detection WaveDetect figurative language detection false-premise detection SH-Detection weakly supervised anomaly detection Voice Activity Detection (VAD)Grad Detect SV-Detect temporal glitch detection Distill to Detect

Recent events (1)

4arXiv · cs.CL·May 28, 2026·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.

Agent and Tool Ecosystem RoBERTa counterfactual data augmentation Polymarket +2 more