The paper presents a method for cross‑architecture knowledge distillation from a fine‑tuned DINOv2 Vision Transformer teacher to a lightweight bidirectional Visual State Space Model (LVSSM) student for tea leaf disease classification. By addressing training‑stability issues with a progressive convolutional stem and gated selective‑scan block, the 4.45 M‑parameter student achieves a mean test accuracy of 95.41%—a 3.09‑point improvement over the teacher’s 92.32%—while using only one‑fifth of the teacher’s parameters. Ablation studies show that simple logit‑level distillation outperforms intermediate feature alignment, and the gains are specific to students that start below the teacher’s performance.
arXiv:2609.07915v1 Announce Type: cross Abstract: Large vision models provide useful representations for remote-sensing segmentation but are often too expensive for deployment at the satellite or fie...
arXiv:2609.20290v1 Announce Type: cross Abstract: Detecting crop disease early is central to sustainable agriculture and food security under United Nations Sustainable Development Goal 2 (Zero Hunger...
arXiv:2606. 14684v1 Announce Type: cross Abstract: Real-time fire classification systems require models that are simultaneously accurate, computationally efficient, and deployable on resource-constrained hardware.
arXiv:2412. 01282v2 Announce Type: replace-cross Abstract: Vision-Language Models (VLMs) bring powerful understanding and reasoning capabilities to multimodal tasks.
The paper presents a unified compression framework for Vision Transformers aimed at on‑device plant disease detection in resource‑constrained agricultural settings. It combines Hessian‑Balanced Adaptive Block Pruning, quantization, and attention‑based knowledge distillation, evaluating each component separately before integrating the best performers into a deployment pipeline. On a chilli disease dataset, the compressed models achieve accuracy comparable to the FP32 baseline while reducing model size by 74‑98 %, and the full pipeline attains a 54.5× size reduction to 6.01 MB with 95.13 % accuracy.
The paper presents a method for distilling a large 300‑million‑parameter geospatial foundation model (Prithvi‑EO‑2.0) into a compact 0.7‑million‑parameter EfficientViT‑B0 student for flood segmentation. By using the teacher to supervise additional unlabeled Sentinel‑2 imagery, the student’s training set expands without new manual labels, achieving competitive performance on Sen1Floods11 and STURM‑Flood while remaining smaller and faster. After quantization, the student runs as a 1.5‑MB INT8 TensorRT engine on a Jetson Xavier NX, processing 512×512 images in 5.57 ms with ~14 MB of memory.
arXiv:2607. 14703v1 Announce Type: cross Abstract: Multiple instance learning (MIL) has become the main paradigm for whole-slide image (WSI) analysis in computational pathology.
arXiv:2608. 15972v1 Announce Type: cross Abstract: Synchronized camera and wireless measurements observe the same scene through different physical channels.
arXiv:2604. 27128v2 Announce Type: replace-cross Abstract: Foundation-model pipelines for individual-level livestock monitoring -- combining open-vocabulary detection, promptable video segmentation, and self-supervised visual embeddings -- have raised the accuracy ceiling of precision livestock farming (PLF), but their GPU memory budgets exceed the envelope of commodity edge accelerators.
arXiv:2606. 07882v1 Announce Type: cross Abstract: Different vision neural networks -- trained to classify, contrast, reconstruct, or match images to text -- should have correspondingly different internal representations.
arXiv:2606. 00928v1 Announce Type: cross Abstract: Multiplexed fluorescence microscopy improves tissue segmentation by providing complementary channels including nuclear (DAPI) and membrane (E-cadherin), that together encode richer spatial context than single-channel imaging alone.
Vision Transformers underperform convolutional networks when training data is scarce, and distilling convolutional inductive biases from a CNN teacher is an effective remedy that leaves the deployed model unchanged. General-purpose feature distillation, however, transfers little in this setting.