Machine Learning Research
More Factual LLMs: FactTune, a method to fine-tune LLMs for factual accuracy without human feedback
Large language models sometimes generate false statements. New work makes them more likely to produce factual output.
Generative AI
Tech & Society
Deepfakes Become Politics as Usual: Deepfakes dominate as India’s election season unfolds.
Synthetic depictions of politicians are taking center stage as the world’s biggest democratic election kicks off.
Tech & Society
India Warns Devs — No Unreliable AI: India advises pre-approval for new AI deployments by tech firms.
India advised major tech companies to seek government approval before they deploy new AI models.
Tech & Society
Anthropic Ups the Ante: Anthropic introduces Claude 3, a new trio of multimodal models.
Anthropic announced a suite of large multimodal models that set new states of the art in key benchmarks.
Tech & Society
Google Tests Generative News Tools: Google funds newsrooms to test AI-powered article generation tools.
Google is paying newsrooms to use a system that helps transform press releases into articles.
Tech & Society
Google Releases Open Source LLMs: All we know about Google's Gemma-7B and Gemma-2B models
Google asserted its open source bona fides with new models. Google released weights for Gemma-7B, an 8.5 billion-parameter large language model intended to run GPUs, and Gemma-2B, a 2.5 billion-parameter version intended for deployment on CPUs and edge devices.
Tech & Society
Context Is Everything: Gemini 1.5 Pro, a leap in multimodal AI amid controversy over v1.0
An update of Google’s flagship multimodal model keeps track of colossal inputs, while an earlier version generated some questionable outputs.
Tech & Society
Generated Video Gets Real(er): OpenAI’s Sora, a new player in text-to-video generation
OpenAI’s new video generator raises the bar for detail and realism in generated videos — but the company released few details about how it built the system.
Machine Learning Research
Better Images, Less Training: Würstchen, a speedy, high-quality image generator
The longer text-to-image models train, the better their output — but the training is costly. Researchers built a system that produced superior images after far less training.
Tech & Society
GPT-4 Biothreat Risk is Low: Study finds GPT-4 no more risky than online search in aiding bioweapon development.
GPT-4 poses negligible additional risk that a malefactor could build a biological weapon, according to a new study. OpenAI compared the ability of GPT-4 and web search to contribute to the creation of a dangerous virus or bacterium. The large language model was barely more helpful than the web.
Tech & Society
High Yields for Small Farms: AI elevates chili farming in India with smarter yields.
Indian farmers used chatbots and computer vision to produce higher yields at lower costs. The state government of Telangana in South India partnered with agricultural aid organization Digital Green to provide AI tools to chili farmers.
Tech & Society
Standard for Media Watermarks: C2PA introduces watermark tech to combat media misinformation.
An alliance of major tech and media companies introduced a watermark designed to distinguish real from fake media starting with images. The Coalition for Content Provenance and Authenticity (C2PA) offers an open standard that marks media files with information about their creation and editing.