NeurIPS 2025: Auto-Compressing Networks for efficient neural representations
Auto-Compressing Networks explore how architectural design can reduce redundancy while retaining useful learned representations.
Machine Learning
ICLR 2026: measuring the hidden geometry of learned representations
Representation holonomy examines how neural features change along input paths, extending analysis beyond pointwise similarity.
TMLR 2026: prescriptive SVD-inspired attention via spectral energy retention
The paper “Prescriptive SVD-Inspired Attention via Spectral Energy Retention” has been published in Transactions on
ICASSP 2026: representation-diverse self-supervision for bioacoustic learning
“Representation-Diverse Self-Supervision for Cross-Domain Bioacoustic Learning in Low-Resource Settings” was presented at ICASSP 2026. The
AISTATS 2026: process-tensor tomography of SGD and training memory
The paper “Process-Tensor Tomography of SGD: Measuring Non-Markovian Memory via Back-Flow of Distinguishability” was published
ICPRAM 2026: two-stage angular alignment for positive-unlabeled learning
“Two-Stage Angular Alignment for Positive-Unlabeled Learning” proposes a geometric weak-supervision method for settings where only
IEEE Access 2026: interpretable Vision Transformers via SVDA
“Interpretable Vision Transformers in Image Classification via SVDA” extends the group’s SVD-inspired attention work to
TMLR: angular regularization for positive-unlabeled learning
The paper “Angular Regularization for Positive-Unlabeled Learning on the Hypersphere” received TMLR certification and was
IEEE Access 2025: geometry meets attention through SVD-inspired transformers
The article “Geometry Meets Attention: Interpretable Transformers via SVD Inspiration” was published in IEEE Access
TPAMI 2024: monocular depth estimation, a thorough review
The article “Monocular Depth Estimation: A Thorough Review” was published in IEEE Transactions on Pattern