The Research Desk.

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

Last Updated: Apr 12, 2026, 11:30 AM PT

AlphaXiv Trending

In-Place Test-Time Training
AlphaXiv
147

In-Place Test-Time Training

Guhao Feng, Shengjie Luo, Kai Hua

#machine-learning#alphaxiv
Neural Computers
AlphaXiv
140

Neural Computers

Mingchen Zhuge, Changsheng Zhao, Haozhe Liu

#alphaxiv
PaperOrchestra: A Multi-Agent Framework for Automated AI Research Paper Writing
AlphaXiv
98

PaperOrchestra: A Multi-Agent Framework for Automated AI Research Paper Writing

Yiwen Song, Tomas Pfister

#reinforcement-learning#retrieval#alphaxiv
Gym-Anything: Turn any Software into an Agent Environment
AlphaXiv
54

Gym-Anything: Turn any Software into an Agent Environment

Pranjal Aggarwal, Graham Neubig, Sean Welleck

#reinforcement-learning#alphaxiv
SkillClaw: Let Skills Evolve Collectively with Agentic Evolver
AlphaXiv
51

SkillClaw: Let Skills Evolve Collectively with Agentic Evolver

Ziyu Ma, Shidong Yang, Yuxiang Ji

#reinforcement-learning#alphaxiv
KnowU-Bench: Towards Interactive, Proactive, and Personalized Mobile Agent Evaluation
AlphaXiv
38

KnowU-Bench: Towards Interactive, Proactive, and Personalized Mobile Agent Evaluation

Tongbo Chen, Zhengxi Lu, Zhan Xu

#reinforcement-learning#alphaxiv

HuggingFace Daily Papers

ClawBench: Can AI Agents Complete Everyday Online Tasks?
HuggingFace
237

ClawBench: Can AI Agents Complete Everyday Online Tasks?

Yuxuan Zhang, Yubo Wang, Yipeng Zhu +2 more

#reinforcement-learning#reacher-z
Training a Student Expert via Semi-Supervised Foundation Model Distillation
HuggingFace
8

Training a Student Expert via Semi-Supervised Foundation Model Distillation

Pardis Taghavi, Tian Liu, Renjie Li +2 more

#machine-learning#efficiency
The Master Key Hypothesis: Unlocking Cross-Model Capability Transfer via Linear Subspace Alignment
HuggingFace
5

The Master Key Hypothesis: Unlocking Cross-Model Capability Transfer via Linear Subspace Alignment

Rishab Balasubramanian, Pin-Jie Lin, Rituraj Sharma +2 more

#safety-alignment
QEIL v2: Heterogeneous Computing for Edge Intelligence via Roofline-Derived Pareto-Optimal Energy Modeling and Multi-Objective Orchestration
HuggingFace
4

QEIL v2: Heterogeneous Computing for Edge Intelligence via Roofline-Derived Pareto-Optimal Energy Modeling and Multi-Objective Orchestration

Satyam Kumar, Saurabh Jha

Appear2Meaning: A Cross-Cultural Benchmark for Structured Cultural Metadata Inference from Images
HuggingFace
4

Appear2Meaning: A Cross-Cultural Benchmark for Structured Cultural Metadata Inference from Images

Yuechen Jiang, Enze Zhang, Md Mohsinul Kabir +2 more

#computer-vision#efficiency#reasoning
CylinderDepth: Cylindrical Spatial Attention for Multi-View Consistent Self-Supervised Surround Depth Estimation
HuggingFace
2

CylinderDepth: Cylindrical Spatial Attention for Multi-View Consistent Self-Supervised Surround Depth Estimation

Samer Abualhanud, Christian Grannemann, Max Mehltretter

#abualhanud