AI
Geometric Evolution Maps: Extracting Stable Concept Probes from Transformer Residual Streams
Researchers have introduced Geometric Evolution Maps (GEMs), a new approach to extracting stable con...
AI
Beyond Recall: Behavioral Specification as an Interpretive Layer for AI Personalization
Researchers introduced a new approach to improve AI personalization by creating an 'interpretive lay...
AI
The Good, the Bad, and the Ugly of Markov Boundary for Tabular Prediction
Researchers investigated the effectiveness of Markov boundary in tabular prediction tasks. They foun...
AI
Can Predicted Dynamics Exist in the Physical World?
Researchers have investigated whether AI-generated predictions about the behavior of objects in the ...
AI
GEM: Geometric Erasure by Contrastive Velocity Matching in Rectified Flows
Researchers have introduced GEM (Geometric Erasure by Contrastive Velocity Match...
AI
MPCoT: Reward-Guided Multi-Path Latent Reasoning for Test-Time Scalable Vision-Language-Action
Researchers have proposed a new framework called MPCoT for improving the perform...
AI
Right Family, Wrong Skill: Benchmarking Risk Exposure in Agent Skill Retrieval
Researchers have created a benchmark to test the reliability of agent skill retr...
AI
A Tool to Map AI Programs in the U.S.: A Snapshot from April 2026 and an Analysis of Requirements for AI Majors and Minors
Researchers have developed a tool to map and analyze AI programs in the United S...