Detecting Chinese Disinformation with Fine–Tuned BERT and Contextual Techniques
Bibliographic record
Abstract
In the digital age, misinformation poses a significant threat to social cohesion, digital platform integrity, and political stability, particularly in countries with large online populations like China. Objectives – We examine how model architecture (BERT versus RoBERTa) and advanced strategies influence accuracy, precision, and recall on the multi – source MCFEND corpus. Building on large language models (LLMs) like BERT (Bidirectional Encoder Representations from Transformers) provides a promising avenue for addressing this challenge. This study presents a novel approach to specifically detecting Chinese misinformation using fine-tuned BERT models, incorporating techniques such as Contextual Unit Obscuration, Multi-span Concealment, and Adaptive Concealment. These methods enhance the models’ ability to capture linguistic nuances and contextual cues specific to Chinese text. Our BERT-based and RoBERTa-based fine-tuned models demonstrate superior performance compared to traditional fine-tuning methods and other state-of-the-art approaches, achieving an accuracy of 83.1% —surpassing state-of-the-art approaches – and achieve notable precision and recall scores over 0.73, marks a significant improvement over many existing detection frameworks. This research supports global efforts to combat misinformation by providing a robust framework across linguistic and cultural contexts. Integrating these models with media literacy and policy initiatives is vital to enhancing digital platform integrity, building a resilient information ecosystem, and promoting informed public discourse.
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 distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.001 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
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 teacher head, 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".