constraint-aware-counterfactual-editing-for-aspect-based-sentiment-analysis-f29829bf·1 events·first seen Aliases: Constraint-Aware Counterfactual Editing for Aspect-Based Sentiment Analysis
Researchers introduce CAVE-ABSA, a framework for generating and validating aspect-level counterfactuals in Aspect-Based Sentiment Analysis (ABSA). The system localizes opinion spans, performs controlled rewriting, and filters candidates using semantic similarity, AMR-guided structural preservation, fluency, and contradiction detection. The framework targets a known weakness in counterfactual evaluation — existing methods flip sentence-level labels but may violate aspect-level constraints — and is designed to support both robustness evaluation and data augmentation for ABSA models.