Papers › What Differentiates Educational Literature? A Multimodal Fusion Approach of...

What Differentiates Educational Literature? A Multimodal Fusion Approach of Transformers and Computational Linguistics

26 Nov 2024arXiv:2411.17593archive 2025-07-28

Jordan J. Bird

The integration of new literature into the English curriculum remains a challenge since educators often lack scalable tools to rapidly evaluate readability and adapt texts for diverse classroom needs. This study proposes to address this gap through a multimodal approach that combines transformer-based text classification with linguistic feature analysis to align texts with UK Key Stages. Eight state-of-the-art Transformers were fine-tuned on segmented text data, with BERT achieving the highest unimodal F1 score of 0.75. In parallel, 500 deep neural network topologies were searched for the classification of linguistic characteristics, achieving an F1 score of 0.392. The fusion of these modalities shows a significant improvement, with every multimodal approach outperforming all unimodal models. In particular, the ELECTRA Transformer fused with the neural network achieved an F1 score of 0.996. Unimodal and multimodal approaches are shown to have statistically significant differences in all validation metrics (accuracy, precision, recall, F1 score) except for inference time. The proposed approach is finally encapsulated in a stakeholder-facing web application, providing non-technical stakeholder access to real-time insights on text complexity, reading difficulty, curriculum alignment, and recommendations for learning age range. The application empowers data-driven decision making and reduces manual workload by integrating AI-based recommendations into lesson planning for English literature.

PaperPDF

Code

No code repository is listed for this paper in the archive or in Syntology's graph.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

Decision MakingText Classificationtext-classification

Datasets

Introduced by this paper, per the archive.

UK Key Stage Readability

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Text Classification UK Key Stage Readability ELECTRA + ANN F1 99.6 #1 of 15 Archive leaderboard report
Text Classification UK Key Stage Readability ERNIE + ANN F1 99.4 #2 of 15 Archive leaderboard report
Text Classification UK Key Stage Readability XLNet + ANN F1 99.2 #3 of 15 Archive leaderboard report
Text Classification UK Key Stage Readability RoBERTa + ANN F1 98.7 #4 of 15 Archive leaderboard report
Text Classification UK Key Stage Readability Longformer + ANN F1 93.9 #5 of 15 Archive leaderboard report
Text Classification UK Key Stage Readability BERT + ANN F1 90.5 #6 of 15 Archive leaderboard report
Text Classification UK Key Stage Readability ALBERT + ANN F1 79.7 #7 of 15 Archive leaderboard report
Text Classification UK Key Stage Readability BERT F1 75 #8 of 15 Archive leaderboard report
Text Classification UK Key Stage Readability DistilBERT F1 74.4 #9 of 15 Archive leaderboard report
Text Classification UK Key Stage Readability Longformer F1 74 #10 of 15 Archive leaderboard report
Text Classification UK Key Stage Readability XLNet F1 74 #11 of 15 Archive leaderboard report
Text Classification UK Key Stage Readability ERNIE F1 73.6 #12 of 15 Archive leaderboard report
Text Classification UK Key Stage Readability RoBERTa F1 73.1 #13 of 15 Archive leaderboard report
Text Classification UK Key Stage Readability ELECTRA F1 71.3 #14 of 15 Archive leaderboard report
Text Classification UK Key Stage Readability ALBERT F1 67.8 #15 of 15 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Methods

ALIGNAbsolute Position EncodingsAdamAttentionAttention DropoutBERTBPEDense ConnectionsDropoutELECTRALabel SmoothingLayer NormalizationLinear LayerLinear Warmup With Linear DecayMulti-Head AttentionPosition-Wise Feed-Forward LayerResidual ConnectionSoftmaxTransformerWeight DecayWordPiece

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections