CoNST: Context-Aware Neural Method for Word Replacement and Sentence Simplification to Improve Text Accessibility
Bibliographic record
Abstract
A primary task in Natural Language Processing (NLP) is text simplification, which aims to create text that is more comprehensible and accessible to a broader audience. Lowering linguistic difficulty and restructuring sentences enhances readability, making it easier for individuals with varying literacy levels and skills. However, many unsupervised lexical simplification methods currently in use primarily focus on individual complex words without considering the surrounding context, which often results in inappropriate substitutions. In this work, we propose a neural-based sentence simplification approach called CoNST and compare it against a rule-based method and other baseline systems, and operate in three sequential stages. In the CoNST neural-based approach, complex word identification is performed using a Bi-LSTM sequential architecture, followed by substitute generation with a BERT model pre-trained for Masked Language Modeling (MLM). Finally, alternatives are ranked to deliver the simplest possible sentence. In contrast, the experimental results show that the neural-based approach achieves a SARI score of 40.24, outperforming the rule-based technique and other baselines.These results demonstrate the advantages of combining advanced neural models for context-aware word replacement with a structured evaluation model. In an era of information overload and diverse literacy needs, this work contributes a robust and effective method for improving linguistic clarity and accessibility.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.002 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.004 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".