用于可解释多特质作文评分的结构化反馈提取统一框架
A Unified Framework to Elicit Structured Feedback for Interpretable Multi-Trait Essay Scoring
浏览论文内容
中文总结 AI 辅助
该研究针对多特质自动作文评分中反馈与评分分离的问题，提出统一框架HiFTS，结合教师LLM提炼反馈并训练学生模型，在新数据集CFMS-34及ASAP++上取得优异评分与反馈效果。
中文摘要 AI 辅助
多特质自动作文评分（AES）需要依据评分标准对相互关联的特质进行推理，而非孤立地预测分数。现有的反馈增强方法常将反馈与评分分离，或独立评估各特质，削弱了分数与反馈的一致性及与评分标准的契合度。我们提出HiFTS，一种统一的自回归框架，在预测特质级和整体分数前生成分层思维链（CoT）反馈。HiFTS从教师大语言模型（LLM）中提炼出基于评分标准的分层思维链反馈，并训练学生模型联合生成反馈与分数。HiFTS进一步采用分组相对策略优化，其复合奖励平衡了分数一致性、校准度、反馈质量及结构有效性。推理时，轻量全局先验提供整体指导，以减少长形式推理过程中的偏差。我们还引入CFMS-34，这是一个含951篇作文的中文多特质AES数据集，标注了整体分数和34个基于评分标准的特质。在CFMS-34和ASAP++上的实验表明，HiFTS在实现优异的整体和特质级评分的同时，能生成连贯、契合评分标准的反馈。
英文摘要
Multi-trait Automated Essay Scoring (AES) requires rubric-grounded reasoning across interdependent traits, rather than isolated score prediction. Existing feedback-enhanced methods often decouple feedback from scoring or assess traits independently, weakening score--feedback consistency and rubric alignment. We propose HiFTS, a unified autoregressive framework that generates hierarchical CoT feedback before predicting trait-level and holistic scores. HiFTS distills rubric-grounded hierarchical CoT feedback from a teacher LLM and trains student models to jointly generate feedback and scores. HiFTS further applies Group Relative Policy Optimization with a composite reward balancing score agreement, calibration, feedback quality, and structural validity. At inference, a lightweight global prior provides holistic guidance to reduce drift during long-form reasoning. We also introduce CFMS-34, a Chinese multi-trait AES dataset with 951 essays annotated with holistic scores and 34 rubric-based traits. Experiments on CFMS-34 and ASAP++ show that HiFTS achieves strong holistic and trait-level scoring while producing coherent, rubric-aligned feedback.
发表机构
- Peking University（北京大学）
- Beijing Normal University（北京师范大学）
机构由 AI 辅助整理，请以论文原文为准。
