资讯

the-decoder2月
Meta Neuroscientist King: "Some of the concepts like reasoning may need to be re-evaluated"
The neuroscientist Jean-Rémi King leads the Brain & AI team in Meta’s AI division. In an interview with The Decoder, he discusses the connection between AI and neuroscience, the challenges of ...
the-decoder10月
Microsoft's Copilot falsely accuses court reporter of crimes he covered
Microsoft Copilot falsely accused court reporter Martin Bernklau of being convicted of child and ward abuse, being involved in a prison escape, and exploiting grieving women as an unscrupulous ...
the-decoder9月
Alibaba's Qwen 2.5 AI models are gunning for Llama 3's crown in latest benchmark
Alibaba has introduced Qwen 2.5, a new series of AI models that are optimized for general language, programming, and mathematics. The models are available in sizes ranging from 0.5 to 72 billion ...
the-decoder2月
Safety assessments show that OpenAI's o3 is probably the company's riskiest AI model to date
OpenAI’s latest language models, o3 and o4-mini, incorporate advanced reasoning capabilities and extensive tool use, including image analysis, Python execution, and web browsing. According to OpenAI, ...
the-decoder1 年
Research shows that high-quality education data is key to AI performance - THE DECODER
To reach the performance of FineWeb-Edu, other datasets like C4 or Dolma need up to 10 times more training data. This again shows the effectiveness of focusing on high quality educational data, ...
the-decoder1 年
OpenAI's new multimodal "GPT-4 omni" combines text, vision, and audio in a single model
The "o" in GPT-4o stands for "omni," reflecting the model's ability to process multiple input and output types through the same neural network. One of the key features of GPT-4o is its impressive ...
the-decoder3月
OpenAI adopts competitor Anthropic's standard for AI data access
OpenAI CEO Sam Altman announced on X that the company will integrate Anthropic's Model Context Protocol (MCP) into its product lineup. MCP is an open-source standard that helps AI models deliver more ...
the-decoder2月
OpenAI’s new o3 and o4-mini models reason with images and tools
Video: OpenAI Setting new benchmark records OpenAI o3, first introduced in December 2024 and refined since then, is reportedly the company's most powerful reasoning model. OpenAI says it demonstrates ...
the-decoder11月
Language models like GPT-4 memorize more than they reason, study finds - THE DECODER
MIT and Boston University researchers conducted a study showing that large language models like GPT-4 perform significantly worse on counterfactual task variants compared to standard tasks, suggesting ...
the-decoder6月
OpenAI rolls out enhanced memory for ChatGPT, allowing it to reference previous chats
OpenAI introduces a new feature for ChatGPT that allows the chatbot to access previous conversations with the user across chats. When starting a new conversation, ChatGPT remembers these previous ...
the-decoder1 年
Eight more publishers sue OpenAI and Microsoft for massive copyright ...
Eight US publishers, including the New York Daily News and the Chicago Tribune, are suing OpenAI and Microsoft for massive copyright infringement. They accuse the AI companies of using millions of ...
the-decoder5月
OpenAI updates ChatGPT with new feature and new GPT-4o model
OpenAI has rolled out an updated version of GPT-4o in ChatGPT, extending its knowledge base through June 2024. The refresh aims to provide more current and contextual responses across topics. The ...