Researchers have developed HiFTS, a novel autoregressive framework designed to improve automated essay scoring (AES) by integrating feedback generation with score prediction. This unified approach aims to enhance score-feedback consistency and better align with scoring rubrics. HiFTS distills hierarchical feedback from a teacher LLM and trains student models to jointly produce feedback and scores, utilizing Group Relative Policy Optimization for a balanced reward system. The framework also introduces CFMS-34, a new Chinese dataset for multi-trait AES, and demonstrates strong performance on both CFMS-34 and the ASAP++ dataset. AI
IMPACT This framework could lead to more accurate and interpretable automated essay scoring systems, potentially impacting educational technology and assessment tools.
RANK_REASON The cluster describes a new research paper detailing a novel framework for automated essay scoring. [lever_c_demoted from research: ic=1 ai=1.0]
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →
