Empirical Comparison of Encoder-Based Language Models and Feature-Based Supervised Machine Learning Approaches to Automated Scoring of Long Essays

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Wang, Kuo, Hua, Haowei, Yan, Pengfei, Jiao, Hong, Song, Dan
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912805768134656
author Wang, Kuo
Hua, Haowei
Yan, Pengfei
Jiao, Hong
Song, Dan
author_facet Wang, Kuo
Hua, Haowei
Yan, Pengfei
Jiao, Hong
Song, Dan
contents Long context may impose challenges for encoder-only language models in text processing, specifically for automated scoring of essays. This study trained several commonly used encoder-based language models for automated scoring of long essays. The performance of these trained models was evaluated and compared with the ensemble models built upon the base language models with a token limit of 512?. The experimented models include BERT-based models (BERT, RoBERTa, DistilBERT, and DeBERTa), ensemble models integrating embeddings from multiple encoder models, and ensemble models of feature-based supervised machine learning models, including Gradient-Boosted Decision Trees, eXtreme Gradient Boosting, and Light Gradient Boosting Machine. We trained, validated, and tested each model on a dataset of 17,307 essays, with an 80%/10%/10% split, and evaluated model performance using Quadratic Weighted Kappa. This study revealed that an ensemble-of-embeddings model that combines multiple pre-trained language model representations with gradient-boosting classifier as the ensemble model significantly outperforms individual language models at scoring long essays.
format Preprint
id arxiv_https___arxiv_org_abs_2601_02659
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Empirical Comparison of Encoder-Based Language Models and Feature-Based Supervised Machine Learning Approaches to Automated Scoring of Long Essays
Wang, Kuo
Hua, Haowei
Yan, Pengfei
Jiao, Hong
Song, Dan
Computation and Language
Machine Learning
Long context may impose challenges for encoder-only language models in text processing, specifically for automated scoring of essays. This study trained several commonly used encoder-based language models for automated scoring of long essays. The performance of these trained models was evaluated and compared with the ensemble models built upon the base language models with a token limit of 512?. The experimented models include BERT-based models (BERT, RoBERTa, DistilBERT, and DeBERTa), ensemble models integrating embeddings from multiple encoder models, and ensemble models of feature-based supervised machine learning models, including Gradient-Boosted Decision Trees, eXtreme Gradient Boosting, and Light Gradient Boosting Machine. We trained, validated, and tested each model on a dataset of 17,307 essays, with an 80%/10%/10% split, and evaluated model performance using Quadratic Weighted Kappa. This study revealed that an ensemble-of-embeddings model that combines multiple pre-trained language model representations with gradient-boosting classifier as the ensemble model significantly outperforms individual language models at scoring long essays.
title Empirical Comparison of Encoder-Based Language Models and Feature-Based Supervised Machine Learning Approaches to Automated Scoring of Long Essays
topic Computation and Language
Machine Learning
url https://arxiv.org/abs/2601.02659