AI Daily 归档 - Rebabel

AI Daily – 2025-06-14(Morning)

AI Anthropic Claude Multi-Agent Research Debate Between Jensen Huang and Anthropic CEO Deutsche Telekom European Industrial AI Cloud Flying Hard Drive Enclosures to Bypass Chip Embargo Industrial AI Cloud Multi-Agent Systems New York State RAISE Act Nvidia RAISE Act Sovereign AI

AI Daily – 2025-06-13(Evening)

AIUI full-duplex interaction technology AMD MI350 Ant Group Ming Lite Omni DeepSeek Nano-vLLM iFlytek AIUI L2 mass-produced vehicles achieving L4 autonomous driving Long-context processing capability of o3-pro model OpenAI o3-pro Performance comparison between AMD MI350X and B200 Tesla Robotaxi Vision-Language-Action model VLA Yuanrong Qixing VLA

AI Daily – 2025-06-13(Morning)

AI Agent AI ethics Autonomous AI Collaborative Intelligence Digital Employee Human-AI Collaboration LLM-HAS LLM-HAS Framework Manus AI Operator AI Agent

AI Daily – 2025-06-12(Evening)

AI Model benchmarking IntPhys 2 benchmark Meta physical reasoning robot control robotics self-supervised learning self-supervised learning pre-training V-JEPA 2 V-JEPA 2 world model World model zero-shot planning

AI Daily – 2025-06-12(Morning)

AI agent observability Databricks Lakebase DPO fine-tuned GPT-4.1 European manufacturing AI cloud platform Meta V-JEPA 2 MLflow 3.0 NVIDIA Industrial AI Cloud Open-source world model for video training OpenAI o3-pro Princeton University HistBench Sakana AI Text-to-LoRA Text-to-LLM adapter

AI Daily – 2025-06-11(Evening)

AGI AI data annotation accuracy 99.7%AI Training Alexandr Wang leads Superintelligence team Data Annotation Meta Meta acquires Scale AI equity Model Accuracy Model training cycle shortened by 40%Scale AI Superintelligence Training data contamination rate reduced

AI Daily – 2025-06-11(Morning)

IBM Magistral Magistral-Small-2506 Military AI application testing o3-pro o3-pro pricing OpenAI Quantum Computer Scale AI investment Starling Quantum Computer Super Intelligence Lab

AI Daily – 2025-06-10(Evening)

AI innovation DeepSeek DeepSeek R1 reasoning model Mistral AI Magistral series multimodal large model multimodal large model human thinking map Open-source model OpenAI OpenAI o4 reinforcement learning training reasoning model Reinforcement learning Xiaohongshu dots.llm1 MoE model

AI Daily – 2025-06-10(Morning)

AI strategy Apple AI privacy protection policy Apple Intelligence Traditional Chinese support Apple WWDC25 Cross-system ecosystem AI integration Foundation framework Generative AI Siri release date On-device AI Siri upgrade System-wide translation Visual intelligent search watchOS Smart Stack feature Xcode Vibe Coding