UniReps @ NeurIPS 2023

59 papers

Top Research Directions

Research directions most represented by UniReps @ NeurIPS 2023 papers.

Comparing Neural Network Representations with Centered Kernel Alignment 4
Comparing Neural Representations Through Relational Geometry 3
Synergistic Multimodal Control for Generative Synthesis 2
Scalable N-Way Model Merging and Integration 1
How Fine Tuning Reshapes Internal Model Circuitry 1
Unifying Self-Supervised Learning via Spectral and Information Theory 1
Generative Hallucination and the Fidelity Gap in Neural Decoding 1
The Emergence of Latent Spatial Geometry in Neural Networks 1
The Convergence of Neural Representations 1
Generalizing Neural Processing Motifs Across Senses and Regions 1
Emergence of Functional Neural Connectivity from Spontaneous Activity 1
Auxiliary Unimodal Supervision for Balanced Multimodal Learning 1
Bidirectional Alignment of Biological and Artificial Intelligence 1
Topographic Mapping in Neural Circuit Architecture 1
The Representational Alignment Gap in Artificial Vision 1

Top Authors

AuthorPapers
Alex H Williams3
Ekdeep Singh Lubana3
Gregor Bachmann3
Lorenzo Noci3
Thomas Hofmann3
Brett W. Larsen2
Carl Allen2
Colin Conwell2
David Krueger2
Fartash Faghri2
Hadi Pouransari2
Hidenori Tanaka2
Max Tegmark2
Mehrdad Farajtabar2
Mohammad Rastegari2
Naoshige Uchida2
Nikola Simidjievski2
Oncel Tuzel2
Raviteja Vemulapalli2
Robert Kirk2

Impactful Papers

Papers from UniReps @ NeurIPS 2023 that shaped research clusters.

Foundational

Influential

Papers

Title
Supervising Variational Autoencoder Latent Representations with Language
Towards Measuring Representational Similarity of Large Language Models
Understanding Learning Dynamics of Neural Representations via Feature Visualization at Scale
Understanding Mode Connectivity via Parameter Space Symmetry
UniCat: Crafting a Stronger Fusion Baseline for Multimodal Re-Identification
Universality of intrinsic dimension of latent representations across models
Variational Classification
What Mechanisms Does Knowledge Distillation Distill?
ZipIt!: Multitask Model Merging without Training

« prev Page 2/2