Should You Fine-Tune BERT for Automated Essay Scoring?
Open accessThe DOI matches its Crossref record; 23 of the 23 references it cites were found in a registry or an index.
Abstract
Most natural language processing research now recommends large Transformer-based models with fine-tuning for supervised classification tasks; older strategies like bag-ofwords features and linear models have fallen out of favor.Here we investigate whether, in automated essay scoring (AES) research, deep neural models are an appropriate technological choice.We find that fine-tuning BERT produces similar performance to classical models at significant additional cost.We argue that while state-of-the-art strategies do match existing best results, they come with opportunity costs in computational resources.We conclude with a review of promising areas for research on student essays where the unique characteristics of Transformers may provide benefits over classical methods to justify the costs.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding
Devlin, J., Chang, M.-W., Lee, K., & Toutanova, K. (2018) · arXiv · doi:10.48550/arxiv.1810.04805 · Cited by 73,173
BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding
Devlin, J., Chang, M.-W., Lee, K., & Toutanova, K. (2019) · Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers) · doi:10.18653/v1/n19-1423 · Cited by 54,325
XLNet: Generalized Autoregressive Pretraining for Language Understanding
Yang, Z., Dai, Z., Yang, Y., Carbonell, J., Salakhutdinov, R., & Le, Q. V. (2019) · arXiv · doi:10.48550/arxiv.1906.08237 · Cited by 6,687
DistilBERT, a distilled version of BERT: smaller, faster, cheaper and lighter
Sanh, V., Debut, L., Chaumond, J., & Wolf, T. (2019) · arXiv · doi:10.48550/arxiv.1910.01108 · Cited by 4,653
Energy and Policy Considerations for Deep Learning in NLP
Strubell, E., Ganesh, A., & McCallum, A. (2019) · Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics · doi:10.18653/v1/p19-1355 · Cited by 3,684 · cited 1×
“• An early deep learning approach using a combination CNN+LSTM architecture that outperformed most reported results at that time (Taghipour and Ng, 2016).”
ResultTransformer-XL: Attentive language models beyond a fixed-length context
Dai, Z., Yang, Z., Yang, Y., Carbonell, J., Le, Q., & Salakhutdinov, R. (2019) · Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics · doi:10.18653/v1/p19-1285 · Cited by 3,169 · cited 2×
“, 2015), these newer models have outperformed the state-of-the-art on numerous tasks, and are currently the most accurate models on a very wide range of tasks (Vaswani et al., 2017; Dai et al., 2019).”
BackgroundAlgorithms of Oppression: How Search Engines Reinforce Racism
Noble, S. U. (2020) · doi:10.2307/j.ctt1pwt9w5 · Cited by 2,709
Parameter-efficient transfer learning for NLP
Houlsby, N., Giurgiu, A., Jastrzebski, S., Morrone, B., de Laroussilhe, Q., Gesmundo, A., Attariyan, M., & Gelly, S. (2019) · arXiv · doi:10.48550/arxiv.1902.00751 · Cited by 2,270
Man is to Computer Programmer as Woman is to Homemaker? Debiasing Word Embeddings
Bolukbasi, T., Chang, K.-W., Zou, J., Saligrama, V., & Kalai, A. (2016) · arXiv · doi:10.48550/arxiv.1607.06520 · Cited by 1,615 · cited 9×
“There is a substantial increase in technical overhead required to implement Transformers and fine-tune them to reach this performance, with minimal gain compared to baselines.”
BackgroundInfluentialA primer in BERTology: What we know about how BERT works
Rogers, A., Kovaleva, O., & Rumshisky, A. (2020) · Transactions of the Association for Computational Linguistics · doi:10.1162/tacl_a_00349 · Cited by 1,552 · cited 4×
“More broadly, the natural language processing (NLP) research community in recent years has been dominated by deep neural network research, in particular, the Transformer architecture popularized by BERT (Devlin et al., 2019).”
BackgroundInfluential
Cited by
Machine learning based feedback on textual student answers in large courses
Bernius, J. P., Krusche, S., & Bruegge, B. (2022) · Computers and Education: Artificial Intelligence · doi:10.1016/j.caeai.2022.100081 · Cited by 75
Towards Trustworthy AutoGrading of Short, Multi-lingual, Multi-type Answers
Schneider, J., Richner, R., & Riser, M. (2023) · International Journal of Artificial Intelligence in Education · doi:10.1007/s40593-022-00289-z · Cited by 48
Henkel, O., Hills, L., Boxer, A., Roberts, B., & Levonian, Z. (2024) · Proceedings of the Eleventh ACM Conference on Learning @ Scale · doi:10.1145/3657604.3664693 · Cited by 42
“Up recently, most SOTA approaches relied primarily on transfer learning and fine-tuning models for specific tasks [10], which models required extensive technical expertise and large datasets [17, 20].”
BackgroundFerreira Mello, R., Pereira Junior, C., Rodrigues, L., Pereira, F. D., Cabral, L., Costa, N., Ramalho, G., & Gasevic, D. (2025) · Proceedings of the 15th International Learning Analytics and Knowledge Conference · doi:10.1145/3706468.3706481 · Cited by 41
A benchmark dataset with human validation for AI-assisted technical answer evaluation
Sheikh, J. K., & Soni, H. K. (2026) · Discover Education · doi:10.1007/s44217-026-01975-z
Metadata
- Authors
- Mayfield, E., & Black, A. W.
- Venue
- Proceedings of the Fifteenth Workshop on Innovative Use of NLP for Building Educational Applications
- Published
- 2020
- Pages
- 151-162
- OpenAlex
- W3037207300
- Cited by
- 137
- References
- 23
- FWCI
- 6.75
- Type
- conference paper
- Language
- en
- Open access
- gold
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
How to cite
Cite CoChat for the checks on this page, and the paper for its own findings.
- This page
- CoChat. (2026). Citation check: Should You Fine-Tune BERT for Automated Essay Scoring? https://app.cochat.ai/papers/10.18653/v1/2020.bea-1.15
- The paper
- Mayfield, E., & Black, A. W. (2020). Should You Fine-Tune BERT for Automated Essay Scoring? Proceedings of the Fifteenth Workshop on Innovative Use of NLP for Building Educational Applications. https://doi.org/10.18653/v1/2020.bea-1.15
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.