The Research Desk.

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

Last Brew Time: Aug 27, 2026, 11:14 AM PT

X.com Research Buzz

Sycophantic Chatbots Cause Delusional Spiraling, Even in Ideal Bayesians
X.com
21423

Sycophantic Chatbots Cause Delusional Spiraling, Even in Ideal Bayesians

Kartik Chandra, Max Kleiman-Weiner, Jonathan Ragan-Kelley, Joshua B. Tenenbaum

MIT CSAIL, University of Washington, Seattle, MIT Department of Brain and Cognitive Sciences

Retrieval
The Tragedy of the Cognitive Commons: How AI Could Disrupt the Regeneration of Professional Expertise
X.com
5315

The Tragedy of the Cognitive Commons: How AI Could Disrupt the Regeneration of Professional Expertise

Nolan Lovett

NATO Special Operations University

AlphaXiv Trending

Human–AI Collaboration at Scale: Task Criticality, Agency, and Friction Across 250,000 Conversations
AlphaXiv
233

Human–AI Collaboration at Scale: Task Criticality, Agency, and Friction Across 250,000 Conversations

Yijia Shao, Dora Zhao, Vishakh Padmakumar

Reinforcement Learning
Prime Agent: A Self-Improving RLM Harness
AlphaXiv
185

Prime Agent: A Self-Improving RLM Harness

Seth Karten, Alex L. Zhang, Kevin Thomas

Affiliation: Princeton University Prime Intellect MIT, Princeton University Prime Intellect MIT

Computer Vision
On-Policy Self-Distillation in Diffusion Models
AlphaXiv
43

On-Policy Self-Distillation in Diffusion Models

Wei Zhou, Xiongwei Zhu, Lingdong Kong

AlphaXiv
29

GLM-5.3-Flash: Frontier Intelligence, Flash Cost

Z.ai

Z.ai

Reinforcement Learning
LeFlow: Generative Latent Flow Planning for World Models
AlphaXiv
28

LeFlow: Generative Latent Flow Planning for World Models

Hsiang-Wei Huang, Jianxu Shangguan, Junbin Lu

Affiliation: University of Washington, United States, University of Washington, United States

Reinforcement Learning
Code World Model: Coding Agent as World Brain
AlphaXiv
27

Code World Model: Coding Agent as World Brain

Yiwen Chen, Guosheng Lin, Chi Zhang

Westlake AGI Lab, AGI Lab, Westlake University, Nanyang Technological University

HuggingFace Daily Papers

Reinforcement Learning
SWE Refactor Bench: Can Coding Agents Complete a Long-Horizon, Whole-Repository Stack Migration?
HuggingFace
13

SWE Refactor Bench: Can Coding Agents Complete a Long-Horizon, Whole-Repository Stack Migration?

Deyao Hong, Yizhe Chi, Wenyi Li, Xiaoqiu Wang, Mingju Gao

Pushing the Limits of High-Resolution Weather Forecasting through Data Scaling
HuggingFace
4

Pushing the Limits of High-Resolution Weather Forecasting through Data Scaling

Yang Zhao, Peisong Niu, Tian Zhou, Ziqing Ma, Guanlong Ma

NJUHML

NLP
Skill Issue: Are Skills Language-Invariant in LLMs?
HuggingFace
3

Skill Issue: Are Skills Language-Invariant in LLMs?

Bobby Cheng, Adam Gaber, Zhengyuan Liu, Catherine Arnett, Omer Goldman

CoLab Lab

Efficiency
Prefix Sliding for efficient test-time scaling
HuggingFace
3

Prefix Sliding for efficient test-time scaling

Niklas Muennighoff, Zhengyang Wang, Zeyi Chen, Weijia Shi, Binyuan Hui

Speech Audio
LibriBrain100: One Hundred Hours of Broad and Deep MEG Data for Neural Speech Decoding at Scale
HuggingFace
2

LibriBrain100: One Hundred Hours of Broad and Deep MEG Data for Neural Speech Decoding at Scale

Francesco Mantegna, Dulhan Jayalath, Gereon Elvers, Tasha Kim, Benjamin Ballyk

Parker Jones Neural Processing Lab (PNPL)

Reinforcement Learning
A Modular Agent for Reliable and Auditable Spatial Relation Verification in CT Scans
HuggingFace
2

A Modular Agent for Reliable and Auditable Spatial Relation Verification in CT Scans

Simon Vincent Abel, Heiko Hillenhagen, Michael Götz, Timo Ropinski, Ayhan Can Erdur