UniReps @ NeurIPS 2025

82 papers

Top Research Directions

Research directions most represented by UniReps @ NeurIPS 2025 papers.

The Real World Validation Gap of Synthetic Data 3
Comparing Neural Representations Through Relational Geometry 2
The Architectural Simplification Gap in Transformer Theory 2
Efficiency and Computational Bottlenecks in Self-Supervised Pretraining 2
Symmetry Invariant Comparison of Neural Representations 1
Geometric Subspace Alignment for Task Transfer 1
Task-Dependent Graph Topology 1
Architectural Inductive Bias of Untrained Neural Networks 1
Unifying Self-Supervised Learning via Spectral and Information Theory 1
Why Standard Training Budgets Underestimate Model Potential 1
The Optimal Layer Depth for Activation Steering 1
The Convergence of Neural Representations 1
Dense Reward Shaping via Latent Embedding Distances 1
Implicit Gradient Descent in Transformer Forward Passes 1
The Single Layer Constraint in Transformer Theory 1

Top Authors

AuthorPapers
Randall Balestriero9
Donato Crisostomi3
Emanuele Rodolà3
Lucas Maes3
Matteo Ferrante3
Meenakshi Khosla3
Alexander S. Ecker2
Ansh Soni2
Carola-Bibiane Schönlieb2
Carolyn Rose2
Christin Seifert2
Damien Scieur2
Fabrizio Silvestri2
Hafez Ghaemi2
Hugues Van Assel2
Jacob A Zavatone-Veth2
Jörg Schlötterer2
Maria Sofia Bucarelli2
Moshe Eliasof2
Nicola Toschi2

Impactful Papers

Papers from UniReps @ NeurIPS 2025 that shaped research clusters.

Foundational

Influential

Papers

Title
Phase codes emerge in recurrent neural networks optimized for modular arithmetic
Quantum Relational Knowledge Distillation
R2-CoD: Understanding Text-Graph Complementarity in Relational Reasoning via Knowledge Co-Distillation
R²-CoD: Understanding Text-Graph Complementarity in Relational Reasoning via Knowledge Co-Distillation
Radial-VCReg: More Informative Representation Learning Through Radial Gaussianization
Relational Representation Learning
Representing Neural Network Layers as Linear Operations via Koopman Operator Theory
Rethinking Objectives for Multi-View and Multi-Modal Contrastive Learning
Self-Supervised Learning from Structural Invariance
SemCLIP: A Semantic Memory-Aligned Vision Language Model
Shared Parameter Subspaces and Cross-Task Linearity in Emergently Misaligned Behavior
Sparse Autoencoder Neural Operators: Model Recovery in Function Spaces
Spectral Insights into Data-Oblivious Critical Layers in Large Language Models
stable-pretraining: Foundation Model Research Made Simple
Stable Single-Pixel Contrastive Learning for Semantic and Geometric Tasks
Superposition disentanglement of neural representations reveals hidden alignment
Superposition in Graph Neural Networks
Symmetry-Aware Fully-Amortized Optimization with Scale Equivariant Graph Metanetworks
The Fragility of Polarity: A Perturbative Analysis of the sign Hypothesis in Sparse Networks
The Performance Cost of Representational Misalignment
Topological Alignment of Shared Vision-Language Embedding Space
Towards Interpretable Deep Neural Networks for Tabular Data
Transformers as Unrolled Inference in Probabilistic Laplacian Eigenmaps
Two-Scale Latent Dynamics for Recurrent-Depth Transformers
Understanding Task Transfer in Vision-Language Models
Unifying Vision-Language Latents for Zero-label Image Caption Enhancement
Universally Converging Representations of Matter Across Scientific Foundation Models
VQ-Kernels: Unraveling Deep Learning of High-Dimensional Data Geometry
Weight Weaving: Parameter Pooling for Data-Free Model Merging
Where does an LLM begin computing an instruction?
Why and How Auxiliary Tasks Improve JEPA Representations
Windsock is Dancing: Adaptive Multimodal Retrieval-Augmented Generation

« prev Page 2/2