The Research Desk.

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

Last Brew Time: Jul 25, 2026, 10:35 AM PT

X.com Research Buzz

NLP
Why are all LLMs Obsessed with Japanese Culture? On the Hidden Cultural and Regional Biases of LLMs
X.com
12556

Why are all LLMs Obsessed with Japanese Culture? On the Hidden Cultural and Regional Biases of LLMs

Joseba Fernandez de Landa, Carla Perez-Almendros, Jose Camacho-Collados

HiTZ Center - Ixa, University of the Basque Country EHU, Cardiff University

Computer Vision
Qwen-Image-3.0
X.com
5462

Qwen-Image-3.0

Alibaba

AlphaXiv Trending

Computer Vision
Masked Visual Actions for Unified World Modeling
AlphaXiv
99

Masked Visual Actions for Unified World Modeling

Hadi Alzayer, Wenlong Huang, Haonan Chen

Stanford University, University of Maryland, College Park, Harvard University

Claude Opus 5 System Card
AlphaXiv
68

Claude Opus 5 System Card

Anthropic

Reinforcement Learning
Measuring Reward-Seeking via Contrastive Belief Updates
AlphaXiv
62

Measuring Reward-Seeking via Contrastive Belief Updates

Axel Højmark, Jérémy Scheurer, EN, Evgenia Nitishinskaya

AlphaXiv
48

SIGReg from First Principles

Reza Bayat

Open Weights and American AI Leadership
AlphaXiv
38

Open Weights and American AI Leadership

Jensen Huang

Reinforcement Learning
AlphaXiv
38

Scaling Agentic RL: 365,000+ Environments for SWE, Terminal, and Search

Daniel Auras, Prime Intellect Team

HuggingFace Daily Papers

Computer Vision
SANA-Video 2.0: Hybrid Linear Attention with Attention Residuals for Efficient Video Generation
HuggingFace
18

SANA-Video 2.0: Hybrid Linear Attention with Attention Residuals for Efficient Video Generation

Junsong Chen, Jincheng Yu, Yitong Li, Shuchen Xue, Haozhe Liu

NVIDIA

Computer Vision
Self-Supervised Learning of Structured Dynamics from Videos
HuggingFace
16

Self-Supervised Learning of Structured Dynamics from Videos

Lukas Knobel, Andrew Zisserman, Yuki M. Asano

Fundamental AI Lab at UTN

Reinforcement Learning
Multi-Turn On-Policy Distillation with Prefix Replay
HuggingFace
8

Multi-Turn On-Policy Distillation with Prefix Replay

Baohao Liao, Hanze Dong, Christof Monz, Xinxing Xu, Li Dong

Microsoft Research

Reinforcement Learning
OpenForgeRL: Train Harness-native Agents in Any Environment
HuggingFace
5

OpenForgeRL: Train Harness-native Agents in Any Environment

Xiao Yu, Baolin Peng, Ruize Xu, Hao Zou, Qianhui Wu

Microsoft

Reinforcement Learning
FinanceComplexQA: Benchmarking Agentic Reasoning on Industrial-grade Financial Documents
HuggingFace
5

FinanceComplexQA: Benchmarking Agentic Reasoning on Industrial-grade Financial Documents

Xianfu Cheng, Shiwei Zhang, Jiyu Zhao, Jian Yang, Xinyuan Wang

Beihang University

Efficiency
Dataset Distillation by Influence Matching
HuggingFace
0

Dataset Distillation by Influence Matching

Haoru Tan, Wang Wang, Sitong Wu, Xiuzhe Wu, Yangtian Sun

Hong Kong University