Apr 23
Exploring Data Augmentation and Resampling Strategies for Transformer-Based Models to Address Class Imbalance in AI Scoring of Scientific Explanations in NGSS Classroom
★★★★★
significance 2/5
This study investigates data augmentation strategies to improve the automated scoring of student scientific explanations using transformer-based models. Researchers tested synthetic data from GPT-4 and lexical-based extraction methods to address class imbalance issues in SciBERT models.
Why it matters
Synthetic data generation via LLMs is becoming a critical lever for refining specialized, high-stakes domain-specific scoring models.
Tags
#transformer #data augmentation #nlp #education #scibertRelated coverage
- Global South OpportunitiesPivotal Research Fellowship 2026 (Q3): AI Safety Research Opportunity - Global South Opportunities
- arXiv cs.AIAn Intelligent Fault Diagnosis Method for General Aviation Aircraft Based on Multi-Fidelity Digital Twin and FMEA Knowledge Enhancement
- arXiv cs.AIPExA: Parallel Exploration Agent for Complex Text-to-SQL
- arXiv cs.AIThe Power of Power Law: Asymmetry Enables Compositional Reasoning
- arXiv cs.AIOn the Existence of an Inverse Solution for Preference-Based Reductions in Argumentation