AI Daily 归档 - Page 5 of 15 - Rebabel

AI Daily – 2025-05-27(Morning)

AI agents AI code generation tools AI inference AI safety and ethics AMD AMD MI300X performance Google Veo 3 video generation large language models Llama 3.1 405B multimodal models Nvidia Open-source models Reinforcement learning

AI Daily – 2025-05-26(Evening)

Agibot EVAC world model AI Agent DeepSeek-V3-0526 DeepSeek-V3-0526 performance benchmark against GPT-4.5 Embodied Intelligence Grok 3 Grok 3 thought pattern identity recognition issue IBM watsonx Orchestrate enterprise-grade AI large language model Multimodal Reinforcement learning Tsinghua RIFLEx video generation duration extension

AI Daily – 2025-05-26(Morning)

AI Agent AI agent architecture AI ethics AI regulation Claude 4 Claude 4 programming capabilities Gemini 2.5 Pro performance Gemini model large language model Multimodal AI Reinforcement learning RLHF fine-tuning techniques Visual language model evaluation

AI Daily – 2025-05-25(Evening)

AI reasoning and instruction following BAGEL-7B-MoT ByteDance open-source AI model computational time and memory trade-off CVE-2025-37899 vulnerability GPT-4o image generation Linux kernel vulnerability MathIF evaluation benchmark MIT computational theory Mixture of Transformer Experts architecture multimodal AI model OpenAI o3

AI Daily – 2025-05-25(Morning)

AI ethics AI model security controls Claude 4 Claude 4 system prompt leak Linux kernel vulnerabilities o3 model discovers Linux vulnerabilities Radish Run Radish Run Robotaxi commercialization text embedding vec2vec text embedding conversion

AI Daily – 2025-05-24(Evening)

AI Agent AI Model Claude 4 Claude Opus 4 coding benchmark coding capability GRPO algorithm Multimodal Pixel Reasoner framework reasoning ability Reinforcement learning TensorRT-LLM optimization VCBench mathematical visual reasoning

AI Daily – 2025-05-24(Morning)

AI Agent AI hardware AI Model chip R&D Claude 4 Claude Opus 4 coding capability Gemini Diffusion GR00T robot dream learning large language model OpenAI acquires io hardware company robot learning text diffusion model generation speed Xiaomi Xuanjie O1 chip performance

AI Daily – 2025-05-23(Evening)

agent AGENTIF benchmark test AI Model ASL-3 safety rating Claude 4 Behavior and Safety Evaluation Report Claude 4 Opus coding capability Multimodal multimodal time-series large model ChatTS safety evaluation Sonnet 4 SWE-bench Verified score

AI Daily – 2025-05-23(Morning)

AI Agent AI agent long-term task processing AI Model AI model memory mechanism AI safety Anthropic Anthropic API Claude 4 Claude 4 safety protection ASL-3 Claude Opus 4 coding capabilities coding model Opus 4 Sonnet 4