Microsoft reportedly building a 500B LLM called MAI-1
According to a report by The Information, Microsoft is working on a 500B parameter LLM called MAI-1 that could take…
Microsoft launches Phi-3 Mini, a tiny but powerful LM
Microsoft launched Phi-3 Mini, a tiny language model that is part of the company’s strategy to develop lightweight, function-specific AI…
LLM agents can autonomously exploit one-day vulnerabilities
University of Illinois Urbana-Champaign (UIUC) researchers found that AI agents powered by GPT-4 can autonomously exploit cybersecurity vulnerabilities. As AI…
ChatGPT predicts the future when you use this clever prompt
Baylor University Department of Economics researchers experimented with ChatGPT to test its ability to predict future events. Their clever prompting…
Google’s Infini-attention gives LLMs “infinite” context
Google researchers developed a technique called Infini-attention, which allows LLMs to handle infinitely long text without increasing compute and memory…
Claude 3 Opus blows all LLMs away in book-length summarization
Researchers published a study comparing the accuracy and quality of summaries that LLMs produce. Claude 3 Opus performed particularly well…
Apple’s ReALM ‘sees’ on-screen visuals better than GPT-4
Apple engineers developed an AI system that resolves complex references to on-screen entities and user conversations. The lightweight model could…
DeepMind developed SAFE, an AI agent to fact-check LLMs
Researchers from DeepMind and Stanford University developed an AI agent that fact-checks LLMs and enables benchmarking of AI model factuality.…
Personalized LLMs are becoming more persuasive than humans
A team of researchers found that once a large language model (LLM) is personalized with a person’s demographic information, it…
Quiet-STaR teaches language models to think before they speak
Researchers from Stanford University and Notbad AI developed Quiet-STaR, a technique that trains a language model (LM) to reason internally…