14 posts were published in the last hour
- 15:34 : Online Monitoring and Corrective Steering of Programming Agents
- 15:34 : MI-MIDI: Mechanistic Interpretability of Text-to-MIDI Generation Models via Probing, Lenses and Steering
- 15:34 : Policy-Masked Private Experts: Auditable and Reversible Capability Access Control in Sparse MoE Models
- 15:34 : Meta AI Releases Muse Glimmer: A 30B Open-Weights Agentic Model That Runs on One Consumer GPU
- 15:34 : SoRoMoX: Fast, Differentiable, and Parallelizable Soft Robot Models
- 15:34 : Newsom Orders AI Cyber Defense Program for California Critical Infrastructure
- 15:34 : Characterizing the Quality Profile of AI-Generated C++ in Production
- 15:3 : Factorized Hypothesis Search for Evidence-to-Taxonomy Retrieval
- 15:3 : Do 3D Medical Foundation Models See Through MRI Artifacts? A Controlled Study of Representation Robustness
- 15:3 : Cryptanalytic Extraction of Isolated Bias-Free GLU Feed-Forward Blocks by Antipodal Separation
- 15:3 : Most Enterprise AI Isn’t Secure. Here’s What Businesses Can Do
- 15:3 : Bypassing Krum: Selection-Aware Backdoor Attacks in Federated Learning
- 15:3 : Mitigating Risk from AI Frontier Models in Healthcare Organizations
- 15:3 : SLED: Scalable Location Encoding via Distillation