TurboVGGT uses adaptive sparse global attention with varying sparsity levels across frames and layers plus frame attention to enable faster multi-view 3D reconstruction while keeping competitive quality versus prior state-of-the-art methods.
hub
An image is worth 16x16 words: Transformers for image recognition at scale
14 Pith papers cite this work. Polarity classification is still indexing.
14
Pith papers citing it
hub tools
citation-role summary
background 1
citation-polarity summary
roles
background 1polarities
background 1representative citing papers
MedCore achieves 60% parameter and 58.4% FLOP reduction on MedSAM with Dice 0.9549 and preserved boundary metrics via dual-intervention pruning and a new boundary leverage principle.
DORA uses an online RL agent to adaptively merge tokens in Vision Transformers, reporting better accuracy-efficiency trade-offs than static baselines on ImageNet and OOD sets.
Unimodal model representations converge to a relational structure captured by the Indra representation via V-enriched Yoneda embedding, which is unique and structure-preserving and improves cross-model and cross-modal robustness when instantiated with angular distance.
WONN is a new oscillatory neural network based on generalized Winfree dynamics that scales competitively to ImageNet-1K and reaches 80.1% accuracy on Maze-hard with 1% of prior model parameters.
NARA introduces a unified self-supervised method for learning relational, context-dependent representations of heterogeneous vector geoentities that improves performance on building classification, traffic prediction, and POI recommendation.
Hierarchy-Aware Cross-Entropy improves image classification by incorporating class hierarchies into the loss through prediction aggregation and ancestral label smoothing, achieving mean accuracy gains of 4.66% in end-to-end training and 2.18% in linear probing.
The work introduces uncertainty-aware foundation models for clinical data by learning set-valued patient representations that enforce consistency across partial observations and integrate multimodal self-supervised objectives.
Replacing softmax attention with entmax-1.5 in DINOv1 ViT-S/16 improves semantic segmentation mIoU on three benchmarks while keeping ImageNet linear-probing accuracy unchanged.
The paper presents stable-worldmodel (swm), a platform with high-performance data layer, modern world model baselines, planning solvers, and extended environments for reproducible research and generalization evaluation.
Matched learning-rate experiments show LoRA retains substantially higher zero-shot transfer (45% vs 11% on EuroSAT, 58% vs 9% on Pets) than Full FT in CLIP adaptation.
BioMamba matches Transformer performance on bioacoustics tasks while using significantly less VRAM.
Z-Score Filtered SAM retains only high absolute Z-score gradient components per layer during the ascent step and reports higher test accuracy than standard SAM on CIFAR and Tiny-ImageNet benchmarks.
citing papers explorer
-
TurboVGGT: Fast Visual Geometry Reconstruction with Adaptive Alternating Attention
TurboVGGT uses adaptive sparse global attention with varying sparsity levels across frames and layers plus frame attention to enable faster multi-view 3D reconstruction while keeping competitive quality versus prior state-of-the-art methods.
-
MedCore: Boundary-Preserving Medical Core Pruning for MedSAM
MedCore achieves 60% parameter and 58.4% FLOP reduction on MedSAM with Dice 0.9549 and preserved boundary metrics via dual-intervention pruning and a new boundary leverage principle.
-
DORA: Dynamic Online Reinforcement Agent for Token Merging in Vision Transformers
DORA uses an online RL agent to adaptively merge tokens in Vision Transformers, reporting better accuracy-efficiency trade-offs than static baselines on ImageNet and OOD sets.
-
The Indra Representation Hypothesis for Multimodal Alignment
Unimodal model representations converge to a relational structure captured by the Indra representation via V-enriched Yoneda embedding, which is unique and structure-preserving and improves cross-model and cross-modal robustness when instantiated with angular distance.
-
Vision Transformers Need Better Token Interaction
Replacing softmax attention with entmax-1.5 in DINOv1 ViT-S/16 improves semantic segmentation mIoU on three benchmarks while keeping ImageNet linear-probing accuracy unchanged.
- From Per-Image Low-Rank to Encoding Mismatch: Rethinking Feature Distillation in Vision Transformers