18 posts were published in the last hour
- 17:33 : Learning from Multimodal Pseudo-Labels for Robust Open-Vocabulary Instance and Panoptic Segmentation
- 17:33 : Record, train, and deploy from one place with Strands Agents, LeRobot, and Hugging Face Storage Buckets
- 17:32 : The Wording Effect: Quantifying Two-Way Drift in LLM Benchmark Performance
- 17:32 : OpenAI hires new CRO as executive shake-up continues
- 17:32 : APEX: Adaptive Expert Prefetching for Memory-Efficient Edge MoE Inference
- 17:32 : Introducing Gemini 3.7 Flash
- 17:32 : Consolidator: Learning Persistent Routed Memory Across Context Boundaries
- 17:32 : Bring your spreadsheet data to life with Sheets canvas
- 17:32 : REOPD: Reliability-Adaptive Reward Extrapolation for On-Policy Distillation
- 17:3 : Is Per-Agent Policy Composition Safe? Rethinking Successor-Feature Transfer in Cooperative Multi-Agent Reinforcement Learning
- 17:3 : GCPO: Diagnosing and Constraining Subspace Geometry in Rollout RL for LLMs
- 17:3 : Amazon Quick Arrives Inside Word, Excel, PowerPoint, and Outlook
- 17:3 : Hybrid-Policy Self-Editing for Composable Unstructured Knowledge Editing
- 17:3 : Previewing Ultrafast mode: GPT-5.6 Sol at up to 14X the speed
- 17:3 : Rubric Dropout: A Simple Way to Mitigate Reward Hacking in Rubric-as-Reward RL
- 17:3 : What We Learned by Reproducing 2,200 papers from ICML
- 17:3 : Low-Interaction-Rank Learning: Unifying Multiplicative Dual-Encoder Heads
- 17:0 : AI News Brief Hourly Summary 2026-08-13 19h : 20 posts