XGD: Explainable AI-Guided Knowledge Distillation with Feature Refinement for Semantic Segmentation
Bibliographic record
Abstract
Semantic segmentation plays a critical role in applications like predictive maintenance and quality inspection but remains challenging to deploy on resource-constrained devices due to their computational demands. This paper introduces Explainable AI (XAI)-Guided Knowledge Distillation (XGD), a method that combines pixel-wise class probability alignment with saliency map refinement guided by XAI. XGD focuses on refining features at the first decoder layer, allowing lightweight student networks to replicate the performance of more complex teacher models. Experiments conducted on the TTPLA, Substation, and Pascal VOC 2012 datasets show that XGD consistently improves segmentation performance, achieving up to a 4.57% increase in mIoU for the DeepLabV3+ student network with ResNet101 backbone. Ablation studies demonstrate the effects of XGD’s distillation losses, while hyperparameter analysis identifies the optimal settings for efficient knowledge transfer. XGD surpasses existing distillation approaches across various models and backbones, providing an effective, resource-efficient, high-performance approach.
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 | 0.001 | 0.003 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.001 | 0.003 |
| Open science | 0.003 | 0.003 |
| Research integrity | 0.001 | 0.003 |
| Insufficient payload (model declined to judge) | 0.005 | 0.002 |
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".