The Research Desk.

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

Last Brew Time: Jun 26, 2026, 7:31 AM PT

X.com Research Buzz

Reasoning
AI Detectors Fail Diverse Student Populations: A Mathematical Framing of Structural Detection Limits
X.com
8928

AI Detectors Fail Diverse Student Populations: A Mathematical Framing of Structural Detection Limits

Nathan Garland

Griffith University

Sakana Fugu
X.com
4780

Sakana Fugu

Sakana AI

AlphaXiv Trending

Reinforcement Learning
Tmax: A simple recipe for terminal agents
AlphaXiv
193

Tmax: A simple recipe for terminal agents

Hamish Ivison, Junjie Oscar Yin, Rulin Shao

Reinforcement Learning
Qwen-AgentWorld: Language World Models for General Agents
AlphaXiv
122

Qwen-AgentWorld: Language World Models for General Agents

Yuxin Zuo, Zikai Xiao, Li Sheng

Unlimited OCR Works
AlphaXiv
94

Unlimited OCR Works

Youyang Yin, Huanhuan Liu, Qunyi Xie

NLP
Tapered Language Models
AlphaXiv
66

Tapered Language Models

Universite de Montreal, Reza Bayat, Ali Behrouz, Aaron Courville

Cornell University

Reinforcement Learning
OpenThoughts-Agent: Data Recipes for Agentic Models
AlphaXiv
50

OpenThoughts-Agent: Data Recipes for Agentic Models

Negin Raoof, Richard Zhuang, Marianna Nezhurina

Stanford University, University of Texas at Austin, Laude Institute, Harvard University & Harvard Medical School, TU Munich & Munich Center for Machine Learning

Robotics
World Value Models for Robotic Manipulation
AlphaXiv
40

World Value Models for Robotic Manipulation

Zhihao Wang, Jianxiong Li, Yu Cui

HuggingFace Daily Papers

Reasoning
JetSpec: Breaking the Scaling Ceiling of Speculative Decoding with Parallel Tree Drafting
HuggingFace
19

JetSpec: Breaking the Scaling Ceiling of Speculative Decoding with Parallel Tree Drafting

Lanxiang Hu, Zhaoxiang Feng, Yulun Wu, Haoran Yuan, Yujie Zhao

Reinforcement Learning
Running the Gauntlet: Re-evaluating the Capabilities of Agents Beyond Familiar Environments
HuggingFace
9

Running the Gauntlet: Re-evaluating the Capabilities of Agents Beyond Familiar Environments

Mykola Vysotskyi, Runqi Lin, Grzegorz Biziel, Michal Zakrzewski, Sebastian Montagna

University of Oxford

NLP
CoffeeBench: Benchmarking Long-Horizon LLM Agents in Heterogeneous Multi-Agent Economies
HuggingFace
4

CoffeeBench: Benchmarking Long-Horizon LLM Agents in Heterogeneous Multi-Agent Economies

Issa Sugiura, Daichi Hattori, Kazuo Araragi, Keita Ogawa, Shota Onose

Computer Vision
LISA: Likelihood Score Alignment for Visual-condition Controllable Generation
HuggingFace
3

LISA: Likelihood Score Alignment for Visual-condition Controllable Generation

Yanghao Wang, Hongxu Chen, Jiazhen Liu, Zhenqi He, Rui Liu

HKUST

Reinforcement Learning
Discretizing Reward Models
HuggingFace
2

Discretizing Reward Models

Vijay Viswanathan, Shiqi Wang, Devamanyu Hazarika, Chirag Nagpal, Tongshuang Wu

AI at Meta

NLP
When Does Combining Language Models Help? A Co-Failure Ceiling on Routing, Voting, and Mixture-of-Agents Across 67 Frontier Models
HuggingFace
1

When Does Combining Language Models Help? A Co-Failure Ceiling on Routing, Voting, and Mixture-of-Agents Across 67 Frontier Models

Josef Chen

Kaikaku