Rebabel - AI News Daily Global

AI Daily – 2025-06-05(Morning)

AI Agent phishing attack AI collaboration AI energy AI mathematics AI Programming AI security AI video generation ChatGPT ChatGPT meeting minutes mode DeepSeek-R1 model update Duolingo AI course expansion Karpathy UI scripted interaction large language model

AI Daily – 2025-06-04(Evening)

AI credibility assessment AI energy consumption AI hardware optimization AI mathematics research AI medical evaluation AI multi-agent systems AI programming tools AI video generation AlphaProof mathematics competition AudioTrust audio large model evaluation DARPA expMath project FrontierMath benchmark testing GUI-Actor visual positioning

AI Daily – 2025-06-04(Morning)

ChatGPT Memory Function ChatGPT Personalized Interaction Improvement Codex Internet Access Function DeepSeek-R1 Complex Reasoning Capability DeepSeek-R1-0528 Diffusion Model GPT-style Model 3.6-bit Memory Language Model Memory Limit MetaAgentX OpenAI Codex SmolVLA Robot Model Suno AI Music Creation Visual Language Action Model

AI Daily – 2025-06-03(Evening)

AI Agent AI commercialization AI Hallucination AI Music Streaming Fraud AI safety AI Trends Report GTA and GLA Attention Mechanism Internet Queen AI Report LawZero AI Safety Design Reinforcement learning SmolVLA Robot Model Vision-Language Model

AI Daily – 2025-06-03(Morning)

AI Agent AI commercialization BitNet v2 Quantization ChatGPT ChatGPT Memory System Computing Power Requirements Darwin Gödel Machine LLM Multimodal Open-source models PlayDiffusion Audio Editing Reinforcement learning Self-Rewarding Training Framework

AI Daily – 2025-06-02(Evening)

AI Agent Framework AlphaEvolve Mathematical Records Cosmos Transfer Physics Simulation Darwin Gödel Machine Self-Improvement DeepMind AlphaEvolve DeepSeek R1 LLM inference optimization MedHELM Medical Evaluation Multimodal Large Language Models NVIDIA Cosmos ProRL Reinforcement Learning Scalability Reinforcement Learning ProRL Sakana AI DGM

AI Daily – 2025-06-01(Evening)

AI-generated CUDA kernels FLUX.1 Kontext image generation GTA and GLA attention mechanisms Huawei's Pangu Ultra MoE model efficient training system MaskSearch pre-training framework Nankai University and UIUC optimize AI search agent efficiency Pangu Ultra MoE model RISEBench evaluation benchmark SearchAgent-X framework Shanghai AI Lab's RISEBench multimodal evaluation Stanford University AI-generated CUDA kernels outperform humans TON selective reasoning framework Tri Dao, creator of Mamba, proposes GTA and GLA attention mechanisms

AI Daily – 2025-06-01(Morning)

AI Agent AI applications AI Model AlphaEvolve math record Artificial intelligence Computing power bottleneck Deep learning FP4 training method Gemini AI agent Grok system prompt large language model Machine learning Sonnet 4.0 table parsing

AI Daily – 2025-05-31(Evening)

AI optimization Code generation CUDA kernels DeepMind mathematical conjecture library Formal mathematics Huawei S-GRPO method Large model inference RISEBench image editing evaluation Stanford AI-generated CUDA kernels Tongyi Lingma AI IDE