Multi-Class Document Classification using LayoutLMv1 and V2
Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
In the age of information explosion, efficient and accurate information retrieval has become a pivotal task across numerous domains, from finance to healthcare and beyond. The LayoutLM model has enhanced the capabilities of existing NLP techniques by enabling them to extract information from complexly structured documents automatically. In this study, we employ a subset of the RVL-CDIP (Ryerson Vision Lab Complex Document Information Processing) dataset, consisting of 400,000 pictures of data organized into 16 groups. Out of the 16 classes, this subset includes all the classes but there was a limit set to 200 images per class which makes the total amount of images as 3200. Accordingly, 1920, 640, and 640 images make up the training, validation, and testing sets. This dataset is a rich collection of documents with diverse structures and content to demonstrate the effectiveness of our proposed method. LayoutLM, with its unique capability to analyze and understand document structures, has been a pivotal component of our methodology. We utilized LayoutLMv2 and version 1 for the purpose of classifying the documents into their respective categories and comparing the results accordingly. Accordingly, the two versions' accuracies are 80.94 and 68.75 percents.
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.
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 | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| 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 it