The Research Desk.

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

Last Brew Time: Oct 10, 2026, 10:55 AM PT

AlphaXiv Trending

Long-WAM: Scaling the Context of World-Action Models
AlphaXiv
83

Long-WAM: Scaling the Context of World-Action Models

1k, Views, 3k

NVIDIA MIT HKU UCSD

Robotics
Embodied Turing Machines: Stateful Code for Robot Recursive Self-Improvement
AlphaXiv
46

Embodied Turing Machines: Stateful Code for Robot Recursive Self-Improvement

Views

Affiliation: Nanyang Technological University Ropedia, Nanyang Technological University

Reinforcement Learning
MiMo-V2.6: Scaling Reinforcement Learning Towards Self-Improvement
AlphaXiv
32

MiMo-V2.6: Scaling Reinforcement Learning Towards Self-Improvement

Views

Reinforcement Learning
AgentGarten: Code Worlds for Evolving Agents
AlphaXiv
27

AgentGarten: Code Worlds for Evolving Agents

Views

https://github.com/MirroS-Lab/AgentGarten, https://mirros-lab.github.io/agent-garten

Reinforcement Learning
Can Jev be Your Q or Policy in Reinforcement Learning?
AlphaXiv
24

Can Jev be Your Q or Policy in Reinforcement Learning?

Views

Affiliation: Shanxi University, Affiliation: Nanjing University, Affiliation: The Chinese University of Hong Kong, Affiliation: Tianjin University, Shanxi University

Computer Vision
Recursive Self-Improvement through Multi-Agent Self-Supervision
AlphaXiv
17

Recursive Self-Improvement through Multi-Agent Self-Supervision

Views

HuggingFace Daily Papers

Reinforcement Learning
Opera: A Verbal Critic Framework for Long-horizon Coding Agents
HuggingFace
10

Opera: A Verbal Critic Framework for Long-horizon Coding Agents

Kai Mei, Zhiyuan Hu, Yutong Dai, Juntao Tan, Yifan Zhang

Salesforce AI Research

TSUITUENYUE
The Lattice of Transition Laws
HuggingFace
1

The Lattice of Transition Laws

T. Y. Tsui, Jiatao Gu, Lingjie Liu

University of Pennsylvania

Reinforcement Learning
Skill Constellations: Tracing the Supply Chain of Agent Skills on GitHub
HuggingFace
1

Skill Constellations: Tracing the Supply Chain of Agent Skills on GitHub

Fahd Seddik

University of British Columbia - Okanagan Campus

Safety Alignment
Predicting Cable Dynamics with Physical Attention Bias
HuggingFace
1

Predicting Cable Dynamics with Physical Attention Bias

Avihai Giuili, Rotem Atari, Avishai Sintov, Maya Bechler-Speicher

Robotics
Behavioral Persistence and Incomplete Functional Transfer of Co-evolved Communication in Evolutionary Robotics
HuggingFace
0

Behavioral Persistence and Incomplete Functional Transfer of Co-evolved Communication in Evolutionary Robotics

Fernando Montes-Gonzalez

Universidad Veracruzana

Computer Vision
Evaluating the Transfer of Co-Evolved Communication from 2D to 3D Simulation
HuggingFace
0

Evaluating the Transfer of Co-Evolved Communication from 2D to 3D Simulation

Fernando Montes-Gonzalez

Universidad Veracruzana

AI Research Papers — October 10, 2026 Edition | Agentic Brew