The Research Desk.

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

Last Updated: Apr 8, 2026, 12:03 PM PT

X.com Research Buzz

Hiding an Ear in Plain Sight: On the Practicality and Implications of Acoustic Eavesdropping with Telecom Fiber Optic Cables
X.com
6269

Hiding an Ear in Plain Sight: On the Practicality and Implications of Acoustic Eavesdropping with Telecom Fiber Optic Cables

Youqian Zhang, Zheng Fang, Huan Wu +3 more

Sycophantic Chatbots Cause Delusional Spiraling, Even in Ideal Bayesians
X.com
3097

Sycophantic Chatbots Cause Delusional Spiraling, Even in Ideal Bayesians

Kartik Chandra, Max Kleiman-Weiner, Jonathan Ragan-Kelley +1 more

AlphaXiv Trending

Meta-Harness: End-to-End Optimization of Model Harnesses
AlphaXiv
389

Meta-Harness: End-to-End Optimization of Model Harnesses

Yoonho Lee, Roshen Nair, Qizheng Zhang

#efficiency#alphaxiv
Embarrassingly Simple Self-Distillation Improves Code Generation
AlphaXiv
206

Embarrassingly Simple Self-Distillation Improves Code Generation

Ruixiang Zhang, Richard He Bai, Huangjie Zheng

#efficiency#reasoning#alphaxiv
SKILL0: In-Context Agentic Reinforcement Learning for Skill Internalization
AlphaXiv
138

SKILL0: In-Context Agentic Reinforcement Learning for Skill Internalization

Zhengxi Lu, Zhengxi Lu, Zhiyuan Yao +1 more

#reinforcement-learning#alphaxiv
Self-Distilled RLVR
AlphaXiv
96

Self-Distilled RLVR

Chenxu Yang, Chuanyu Qin, Qingyi Si

#efficiency#alphaxiv
CORAL: Towards Autonomous Multi-Agent Evolution for Open-Ended Discovery
AlphaXiv
72

CORAL: Towards Autonomous Multi-Agent Evolution for Open-Ended Discovery

Ao Qu, Ao Qu, Han Zheng +1 more

#reinforcement-learning#robotics#alphaxiv
Vero: An Open RL Recipe for General Visual Reasoning
AlphaXiv
31

Vero: An Open RL Recipe for General Visual Reasoning

Gabriel Sarch, Linrong Cai, Qunzhong Wang

#computer-vision#reinforcement-learning#reasoning#alphaxiv

HuggingFace Daily Papers

General Multimodal Protein Design Enables DNA-Encoding of Chemistry
HuggingFace
17

General Multimodal Protein Design Enables DNA-Encoding of Chemistry

Jarrid Rector-Brooks, Théophile Lambert, Marta Skreta +2 more

#reasoning#multimodal#DISCO-design
ClawsBench: Evaluating Capability and Safety of LLM Productivity Agents in Simulated Workspaces
HuggingFace
10

ClawsBench: Evaluating Capability and Safety of LLM Productivity Agents in Simulated Workspaces

Xiangyi Li, Kyoung Whan Choe, Yimin Liu +2 more

#nlp#reinforcement-learning#safety-alignment#benchflow-ai
Squeez: Task-Conditioned Tool-Output Pruning for Coding Agents
HuggingFace
4

Squeez: Task-Conditioned Tool-Output Pruning for Coding Agents

Ádám Kovács

#reinforcement-learning#efficiency#reasoning#KRLabsOrg
Context-Value-Action Architecture for Value-Driven Large Language Model Agents
HuggingFace
3

Context-Value-Action Architecture for Value-Driven Large Language Model Agents

TianZe Zhang, Sirui Sun, Yuhang Xie +2 more

#nlp#reinforcement-learning
Expert-Choice Routing Enables Adaptive Computation in Diffusion Language Models
HuggingFace
2

Expert-Choice Routing Enables Adaptive Computation in Diffusion Language Models

Shuibai Zhang, Caspian Zhuang, Chihan Cui +2 more

#nlp#computer-vision#zhangshuibai
CUE-R: Beyond the Final Answer in Retrieval-Augmented Generation
HuggingFace
1

CUE-R: Beyond the Final Answer in Retrieval-Augmented Generation

Siddharth Jain, Venkat Narayan Vedam

#retrieval#jainsid24