LEAF distills teacher-aligned student embedding models that achieve new SOTA results on BEIR and MTEB for their size class while requiring only modest data and compute.
Kaggle (2025)
3 Pith papers cite this work. Polarity classification is still indexing.
citation-role summary
citation-polarity summary
roles
dataset 1polarities
use dataset 1representative citing papers
Integrating 15% logistic chaos perturbations into prototypical networks raises 4-way 5-shot brain tumor accuracy to 84.52% by producing more stable, noise-invariant embeddings.
A multi-dataset cross-domain knowledge distillation approach improves unified performance on medical image segmentation, classification, and detection by transferring domain-invariant features from a joint teacher model to task-specific students.
citing papers explorer
-
LEAF: Knowledge Distillation of Text Embedding Models with Teacher-Aligned Representations
LEAF distills teacher-aligned student embedding models that achieve new SOTA results on BEIR and MTEB for their size class while requiring only modest data and compute.
-
Chaos-Enhanced Prototypical Networks for Few-Shot Medical Image Classification
Integrating 15% logistic chaos perturbations into prototypical networks raises 4-way 5-shot brain tumor accuracy to 84.52% by producing more stable, noise-invariant embeddings.
-
Multi-Dataset Cross-Domain Knowledge Distillation for Unified Medical Image Segmentation, Classification, and Detection
A multi-dataset cross-domain knowledge distillation approach improves unified performance on medical image segmentation, classification, and detection by transferring domain-invariant features from a joint teacher model to task-specific students.