TabReformer: Unsupervised Representation Learning for Erroneous Data Detection
Bibliographic record
Abstract
Error detection is a crucial preliminary phase in any data analytics pipeline. Existing error detection techniques typically target specific types of errors. Moreover, most of these detection models either require user-defined rules or ample hand-labeled training examples. Therefore, in this article, we present TabReformer, a model that learns bidirectional encoder representations for tabular data. The proposed model consists of two main phases. In the first phase, TabReformer follows encoder architecture with multiple self-attention layers to model the dependencies between cells and capture tuple-level representations. Also, the model utilizes a Gaussian Error Linear Unit activation function with the Masked Data Model objective to achieve deeper probabilistic understanding. In the second phase, the model parameters are fine-tuned for the task of erroneous data detection. The model applies a data augmentation module to generate more erroneous examples to represent the minority class. The experimental evaluation considers a wide range of databases with different types of errors and distributions. The empirical results show that our solution can enhance the recall values by 32.95% on average compared with state-of-the-art techniques while reducing the manual effort by up to 48.86%.
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 | 0.003 | 0.010 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.002 | 0.002 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.001 | 0.003 |
| Open science | 0.003 | 0.002 |
| Research integrity | 0.001 | 0.002 |
| Insufficient payload (model declined to judge) | 0.004 | 0.003 |
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".