paper-with-me

홈 › Papers

Teacher Network Calibration Improves Cross-Quality Knowledge Distillation

2023-04-15 · Pia Čuk, Robin Senge, Mikko Lauri, Simone Frintrop

We investigate cross-quality knowledge distillation (CQKD), a knowledge distillation method where knowledge from a teacher network trained with full-resolution images is transferred to a student network that takes as input low-resolution images. As image size is a deciding factor for the computational load of computer vision applications, CQKD notably reduces the requirements by only using the student network at inference time. Our experimental results show that CQKD outperforms supervised learning in large-scale image classification problems. We also highlight the importance of calibrating neural networks: we show that with higher temperature smoothing of the teacher's output distribution, the student distribution exhibits a higher entropy, which leads to both, a lower calibration error and a higher network accuracy.

📄 PDF Abstract BibTeX arXiv:2304.07593

Code (1)

piacuk/distillistic 공식 구현 pytorch

Tasks

image-classificationImage ClassificationKnowledge Distillation

Methods 이 논문이 사용한 방법론

Knowledge Distillation A very simple way to improve the performance of almost any machine learning algorithm is to train many different models on the same data and then to average their predictions.…

Similar Papers 제목 키워드 기반

Hard Gate Knowledge Distillation -- Leverage Calibration for Robust and Reliable Language Model

2022-10-22 · Dongkyu Lee, Zhiliang Tian, Yingxiu Zhao, Ka Chun Cheung 외

In knowledge distillation, a student model is trained with supervisions from both knowledge from a teacher and observations drawn from a training data distribution. Knowledge of a teacher is considered a subject that hol…

Knowledge DistillationLanguage ModelingLanguage ModellingSentence+1

The Role of Teacher Calibration in Knowledge Distillation

2025-08-27 · Suyoung Kim, Seonguk Park, Junhoo Lee, Nojun Kwak arxiv

Knowledge Distillation (KD) has emerged as an effective model compression technique in deep learning, enabling the transfer of knowledge from a large teacher model to a compact student model. While KD has demonstrated si…

Knowledge DistillationModel Compression

Reliability Gated Multi-Teacher Distillation for Low Resource Abstractive Summarization

2026-04-03 · Dipto Sumit, Ankan Kumar Roy, Sadia Khair Rodela, Atia Haque Asha 외 arxiv

We study multiteacher knowledge distillation for low resource abstractive summarization from a reliability aware perspective. We introduce EWAD (Entropy Weighted Agreement Aware Distillation), a token level mechanism tha…

Knowledge DistillationSemantic Similarity

Beyond Dark Knowledge: Mixup-Based Distillation for Reliable Predictions

2026-06-10 · José Medina, Paul Honeine, Abdelaziz Bensrhair, Amnir Hadachi arxiv

Knowledge Distillation (KD) and mixup have proven effective at inducing smoothness in class boundaries; KD captures inherent class relationships in probability distributions, and mixup enforces them through convex combin…

Knowledge Distillation

Post-Training Probability Manifold Correction via Structured SVD Pruning and Self-Referential Distillation

2026-01-30 · Aaron R. Flouro, Shawn P. Chadwick arxiv

Large language models are expensive to deploy. We introduce Sparse Knowledge Distillation (SparseKD), a post-training method that compresses transformer models by combining structured SVD pruning with self-referential kn…

Knowledge Distillation