arXiv subjects

Duy Anh Nguyen

Publications and source records attributed to Duy Anh Nguyen.

3 recordsLinked to original sources

arXiv2026

Error-Supervised Synthetic Learner Writing for Automated Essay Scoring

Synthetic essays can help reduce dependence on human-written data in Automated Essay Scoring (AES). However, they often lack realistic errors, limiting their ability to represent authentic human writing, particularly when the target texts are intended to resemble those produced by language learners. In this study, we present a simple approach that introduces error supervision into synthetic essay generation. Specifically, we fine-tune an LLM generator on error-annotated texts of the kind commonly used in Grammatical Error Detection (GED). To assess the utility of the proposed approach, we fine-tune and evaluate AES scorers under three data conditions: authentic essays, synthetic essays generated conventionally, and synthetic essays generated using our proposed approach. The results show that in the larger-data settings, the proposed approach outperforms the conventional synthetic baseline in 11 out of 12 dataset-metric comparisons, with performance in some cases approaching that of models trained on authentic essays. Despite these gains, performance under extremely low-resource settings remains mixed, with advantages over the conventional baseline only becoming more apparent at 200 training essays, although not consistently across datasets. Qualitative and quantitative analyses further show that the proposed approach produces learner-like errors whose distributions broadly resemble those observed in authentic essays.

cs.CL↗

arXiv2026

Does Continued Pretraining on a Learner Corpus Improve Automated Essay Scoring on English Proficiency Tests? Evidence from EFCAMDAT

Automated Essay Scoring (AES) for English proficiency assessment increasingly relies on pretrained transformer models, yet these models are typically trained on general-domain English and may under-represent second-language learner writing. This study investigates whether domain-adaptive continued pretraining (DAPT) on a learner-writing corpus improves transformer-based AES for English proficiency assessment. We perform DAPT on BERT, RoBERTa, and DistilBERT using the EFCAMDAT corpus, then compare the adapted models with their original checkpoints on two English proficiency test datasets, FCE and IELTS, in both in-domain scoring and few-shot cross-dataset transfer. Full-corpus DAPT produces mixed effects across models, datasets, and metrics. Subsequent lexical and syntactic analyses suggest mismatches between EFCAMDAT and the downstream datasets in proficiency level, genre, and communicative purpose. We therefore repeat DAPT using proficiency-specific EFCAMDAT subsets across all three encoder architectures. Proficiency-specific DAPT frequently outperforms full-corpus DAPT and, in some settings, even the non-adapted baseline. Overall, continued pretraining on learner writing can improve in-domain AES, but its benefits depend on both the proficiency composition of the pretraining data and the underlying encoder architecture, and do not consistently extend to cross-test transfer.

cs.CL↗

arXiv2021

Workload Balancing Among Heathcare Workers Under Uncertain Service Time Using Distributionally Robust Optimization

Healthcare systems are facing serious challenges in balancing their human resources to cope with volatile service demand, while at the same time providing necessary job satisfaction to the healthcare workers. We propose in this paper a distributionally robust optimization formulation to generate a task assignment plan that promotes the fairness in allocation, attained by reducing the difference in the total working time among workers, under uncertain service time. The proposed joint chance constraint model is conservatively approximated by a worst-case Conditional Value-at-Risk, and we devise a sequential algorithm to solve the finite-dimensional reformulations which are linear (mixed-binary) optimization problems. We also provide explicit formula in the situation where the support set of the random vectors is a hyperrectangle. The experiment with synthetic data suggests promising results for our approach.

math.OC↗