An Active Learning Pipeline Powered by Image Synthetization and Retrieval Techniques: A Novel Training Approach for Construction-Centric DNNs
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
Recognizing synthetic data’s role in advancing deep learning, this study explores its use in active learning, where systems autonomously identify and rectify their weaknesses. We introduce a novel synthetic oracle pipeline, integrating a Content-Based Image Retrieval (CBIR) module with an established image synthetization module (BlendCon) to create a synthetic oracle. To realize a synthetic-based active learning, the development of a high-performant synthetic oracle is a must. This oracle is crucial for replicating the detected model’s failures and refreshing training data, enhancing active learning effectiveness. We present a prototype of a synthetic oracle, demonstrating proficient image retrieval, generation, and visual quality verification. Our future work will focus on the development of the synthetic oracle and how it can be an enabler to the realization of synthetic-based deep active learning. Our visual verifications demonstrate the efficacy of the synthetic oracle and open novel avenues of research.
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 | 0.001 | 0.001 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.001 | 0.002 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.000 | 0.001 |
| 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