CUBE encodes 3D faces via a grid of learned high-dimensional B-spline features that map parametrically to a base shape plus MLP-refined displacements, enabling dense correspondence and state-of-the-art registration from point clouds or images.
Xcit: Cross-covariance image transformers
6 Pith papers cite this work, alongside 234 external citations. Polarity classification is still indexing.
6
Pith papers citing it
234
external citations · external index
citation-role summary
method 1
citation-polarity summary
years
2026 6roles
method 1polarities
use method 1representative citing papers
Trained correlated-photon illumination paired with a Transformer backend improves object classification accuracy by up to 15 percentage points in photon-starved noisy imaging.
TextTeacher uses frozen text embeddings from captions as semantic anchors to guide vision model training, improving ImageNet accuracy by up to 2.7 p.p. and transfer performance by 1.0 p.p. on average.
The paper casts the standard Transformer block with RoPE as a first-order approximation of a radial–tangential state estimator and introduces a Polar Transformer variant that retains the discarded geometric corrections.
ShapeY is a benchmark dataset and nearest-neighbor protocol that measures shape-based recognition in vision models, revealing that even state-of-the-art networks fail to generalize consistently across 3D viewpoints and non-shape appearance changes.
Non-linear transformers enable cross-domain generalization in in-context RL by representing value functions from different domains with shared weights inside a shared RKHS.
citing papers explorer
-
Representing 3D Faces with Learnable B-Spline Volumes
CUBE encodes 3D faces via a grid of learned high-dimensional B-spline features that map parametrically to a base shape plus MLP-refined displacements, enabling dense correspondence and state-of-the-art registration from point clouds or images.
-
Ultra-low-light computer vision using trained photon correlations
Trained correlated-photon illumination paired with a Transformer backend improves object classification accuracy by up to 15 percentage points in photon-starved noisy imaging.
-
TextTeacher: What Can Language Teach About Images?
TextTeacher uses frozen text embeddings from captions as semantic anchors to guide vision model training, improving ImageNet accuracy by up to 2.7 p.p. and transfer performance by 1.0 p.p. on average.
-
The Transformer as a Polar State Estimator
The paper casts the standard Transformer block with RoPE as a first-order approximation of a radial–tangential state estimator and introduces a Polar Transformer variant that retains the discarded geometric corrections.
-
ShapeY: A Principled Framework for Measuring Shape Recognition Capacity via Nearest-Neighbor Matching
ShapeY is a benchmark dataset and nearest-neighbor protocol that measures shape-based recognition in vision models, revealing that even state-of-the-art networks fail to generalize consistently across 3D viewpoints and non-shape appearance changes.
-
One for All: A Non-Linear Transformer can Enable Cross-Domain Generalization for In-Context Reinforcement Learning
Non-linear transformers enable cross-domain generalization in in-context RL by representing value functions from different domains with shared weights inside a shared RKHS.