The Research Desk.

The most upvoted and starred AI research crossing the community today.

Last Brew Time: Nov 2, 2026, 11:38 AM PT

AlphaXiv Trending

Proving and Improving the Platonic Representation Hypothesis with JEPAs
AlphaXiv
306

Proving and Improving the Platonic Representation Hypothesis with JEPAs

12k, Views

Affiliation: Brown University, Affiliation: New York University, Brown University, New York University

Machine Learning
Training Foundation Models with a Single Algorithm
AlphaXiv
159

Training Foundation Models with a Single Algorithm

2k, Views

Training Foundation Models with a Single Algorithm, Carnegie Mellon University, Johns Hopkins University

Machine Learning
Training language models in simulation for psychiatric intake
AlphaXiv
73

Training language models in simulation for psychiatric intake

2k, Views

Affiliation: Google DeepMind, Affiliation: Google Research, Affiliation: Google for Health, Affiliation: Stanford University, Affiliation: University of California, Los Angeles

Computer Vision
Probing Spatial State Representations in Visual Foundation Models
AlphaXiv
52

Probing Spatial State Representations in Visual Foundation Models

Views

New York University, Nanyang Technological University, The Hong Kong University of Science and Technology, The Chinese University of Hong Kong, Keywords]visual foundation models, spatial state

Hierarchical World Models with Implicit Dynamics
AlphaXiv
18

Hierarchical World Models with Implicit Dynamics

Views

Affiliation: 1 Courant Institute, New York University, Courant Institute, New York University, Meta-FAIR

Robotics
FoldNet++: a Large-Scale Synthetic Dataset for Robotic T-Shirt Folding and Unfolding
AlphaXiv
12

FoldNet++: a Large-Scale Synthetic Dataset for Robotic T-Shirt Folding and Unfolding

Affiliation: CFCS, Peking University, China, CFCS, Peking University, China

HuggingFace Daily Papers

PolyU VCLab
OSWAM: An Effective World Action Model with One-Step Action Generation
HuggingFace
23

OSWAM: An Effective World Action Model with One-Step Action Generation

Guowen Zhang, Liyi Chen, Xu Hu, Pengfei Wang, Jingfan Chen

The Hong Kong Polytechnic University

Machine Learning
Follow the Gradient: Tracing Safety Degradation Back to Training Data
HuggingFace
14

Follow the Gradient: Tracing Safety Degradation Back to Training Data

Saket Reddy, Amro Abbas, Anshuman Suri, Ari Morcos, David Schwab

DatologyAI

Reasoning
SOMA: A Benchmark for Fine-Grained Egocentric Behavior and Body-Motion Understanding
HuggingFace
3

SOMA: A Benchmark for Fine-Grained Egocentric Behavior and Body-Motion Understanding

Matea Tashkovska, Naël Ouerghemi, Haozhe Qi, Alexander Mathis

Mathis Group @ EPFL

Reinforcement Learning
GEPA-Curator: Curated Learning for Self Improving Agents
HuggingFace
2

GEPA-Curator: Curated Learning for Self Improving Agents

Zack Yu, Nicholas Lee, Minjae Lee, Lakshya Agarwal, Luca Manolache

UC Berkeley

Reinforcement Learning
rTaste: Do Research Agents Have Taste? Evaluating Early Stage Research Judgment
HuggingFace
2

rTaste: Do Research Agents Have Taste? Evaluating Early Stage Research Judgment

Art Kanke

Katastrofh
AuditFL: A Contract-Backed Clearing Layer for Open Federated Learning
HuggingFace
1

AuditFL: A Contract-Backed Clearing Layer for Open Federated Learning

Alexandros Balla, Spyros Lalis

HKUST