Computer-Using Agent
AI 速览 OpenAI发布Computer-Using Agent（CUA），一种可像人类一样操作计算机界面的AI代理模型。
人工智能
围绕「人工智能」分类汇集的全球近期技术动态与前沿报道。
Mastering Long Contexts in LLMs with KVPress
AI 速览 KVPress工具可帮助掌握大语言模型的长上下文处理能力。
SmolVLM Grows Smaller – Introducing the 256M & 500M Models!
AI 速览 Hugging Face发布SmolVLM系列新增256M和500M参数模型，进一步缩小视觉语言模型规模。
Hugging Face and FriendliAI partner to supercharge model deployment on the Hub
AI 速览 Hugging Face与FriendliAI达成合作，旨在加速Hub平台上的模型部署。
Introducing multi-backends (TRT-LLM, vLLM) support for Text Generation Inference
AI 速览 Text Generation Inference新增TRT-LLM和vLLM多后端支持。
Timm ❤️ Transformers: Use any timm model with transformers
AI 速览 Hugging Face的timm库现已与Transformers集成，用户可在Transformers中直接使用任意timm模型。
Partnering with Axios expands OpenAI’s work with the news industry
Publishers representing hundreds of newsrooms and content brands are using OpenAI partnerships and grant programs to adopt AI tools and strengthen the news ecosystem, while ChatGPT users gain access to information from leading, reliable publications.
Train 400x faster Static Embedding Models with Sentence Transformers
AI 速览 Sentence Transformers 现可训练静态嵌入模型，速度提升400倍。
AI Agents Are Here. What Now?
AI 速览 AI Agents已到来，下一步待探索。
CO₂ Emissions and Models Performance: Insights from the Open LLM Leaderboard
AI 速览 Hugging Face发布Open LLM Leaderboard分析，探讨大语言模型性能与CO₂排放的关系。
Introducing smolagents: simple agents that write actions in code.
AI 速览 Hugging Face推出smolagents，一种直接在代码中编写动作的简单智能体框架。
Controlling Language Model Generation with NVIDIA's LogitsProcessorZoo
AI 速览 NVIDIA发布LogitsProcessorZoo，用于控制语言模型生成过程。
Deliberative alignment: reasoning enables safer language models
Deliberative alignment: reasoning enables safer language models Introducing our new alignment strategy for o1 models, which are directly taught safety specifications and how to reason over them.
Bamba: Inference-Efficient Hybrid Mamba2 Model
AI 速览 Bamba是一种基于Mamba2架构的混合模型，旨在提升推理效率。
Benchmarking Language Model Performance on 5th Gen Xeon at GCP
AI 速览 在GCP上对第五代Xeon处理器的语言模型性能进行基准测试。
Welcome to the Falcon 3 Family of Open Models!
AI 速览 Falcon 3系列开放模型正式发布。
Sora is here
Our video generation model, Sora, is now available to use at sora.com. Users can generate videos up to 1080p resolution, up to 20 sec long, and in widescreen, vertical or square aspect ratios. You can bring your own assets to extend, remix, and blend, or generate entirely new content from text.
Hugging Face models in Amazon Bedrock
AI 速览 Hugging Face模型现已接入Amazon Bedrock。
Put AI to work for your product team
Put AI to work for your product team
Sora System Card
Sora is OpenAI’s video generation model, designed to take text, image, and video inputs and generate a new video as an output. Sora builds on learnings from DALL-E and GPT models, and is designed to give people expanded tools for storytelling and creative expression.
Introducing ChatGPT Pro
Broadening usage of frontier AI
How good are LLMs at fixing their mistakes? A chatbot arena experiment with Keras and TPUs
AI 速览 研究通过Keras和TPU上的聊天机器人竞技场实验，评估大语言模型修复自身错误的能力。
Welcome PaliGemma 2 – New vision language models by Google
AI 速览 Google发布新一代视觉语言模型PaliGemma 2。
Morgan Stanley is shaping the future of financial services
Morgan Stanley uses AI evals to shape the future of financial services
Rethinking LLM Evaluation with 3C3H: AraGen Benchmark and Leaderboard
AI 速览 研究提出3C3H评估框架及AraGen基准与排行榜，旨在重新审视大语言模型评测方法。
Investing in Performance: Fine-tune small models with LLM insights - a CFM case study
AI 速览 CFM案例研究提出利用LLM洞察微调小模型的方法，以提升小模型性能。
Open Source Developers Guide to the EU AI Act
AI 速览 开源开发者需关注欧盟《人工智能法案》对开源AI系统的合规要求与责任界定。
SmolVLM - small yet mighty Vision Language Model
AI 速览 Hugging Face发布SmolVLM视觉语言模型，主打小尺寸高性能。
Advancing red teaming with people and AI
Advancing red teaming with people and AI
Building smarter maps with GPT-4o vision fine-tuning
Building smarter maps with GPT-4o vision fine-tuning