Greg Loughnane and Chris Alexiuk will discuss how to use hierarchical embeddings to return relevant context from a large dataset, as measured by context precision and context recall via the RAG ASsessment framework, and also show the impact of fine-tuning the embedding model on retrieval metrics.
Anthropic is leveling up in the generative AI race by partnering with Amazon. The company behind Claude committed to expanding its support of the Amazon Bedrock API, and to primarily using AWS as its cloud provider. Amazon will get a minority stake in Anthropic for up to $4 billion in funding.
Kolena, a startup working on a platform that can test and validate solutions for any ML problem, just raised $15M in a funding round led by Lobby Capital. The platform will feature private, fine-grained, customizable testing that adapts to each user's needs.
OpenAI announced it will soon roll out image recognition and speech synthesis features for ChatGPT. These features will be accessible to Enterprise and Pro users. Whether you need a bedtime story or help analyzing a chart, ChatGPT's newly expanded abilities will be here to assist.
Mistral AI is planning to become a leader in the open generative AI community. It has ambitious plans to bridge the gap between proprietary and open-source AI solutions by releasing fully open-source models with a performance comparable to proprietary ones.
Capsule is an AI-powered, browser-based video editor that aims to make video editing easier, faster, and less intimidating. With its user-friendly interface, AI-powered assistant, and soon-to-be-announced collaborative tools, Capsule is definitely a game-changer.
LLM Studio is Galileo’s latest offering for LLM training, evaluation, and monitoring. Consisting of three modules responding to the most common problems faced by developers, Galileo aims to make LLM Studio the all-in-one solution for model evaluation.
OpenAI is announcing the upcoming availability of DALL-E 3, which can boast improved image quality and accuracy following detailed descriptions along with a number of security and privacy measures aimed at mitigation of risks such as propaganda, misinformation, and copyright infringement.
Google has announced three exciting new Bard features, making English conversations even more useful: workspace Extensions which integrate Bard into Google’s app ecosystem, improvements to the ‘Google it’ feature, and the possibility to continue publicly shared conversations.
Anyscale releases Ray 2.7, bringing exciting new features and improvements that make it easy to scale AI and Python workloads — from train and tuning to model serving. Among the most noteworthy new features is the general availability of Ray AI libraries, KubeRay, and RayLLM.
Patronus AI launched on Sept. 14 after securing a 3M seed round led by Lightspeed Venture Partners. The groundbreaking platform aims to help enterprises deploy and scale AI solutions with safety and confidence by automating scoring, test generation, and benchmarking for enterprise AI products.
A team of MIT engineers looks to make clinical evaluation of cerebral palsy patients easier and remotely accessible. They’ve developed a method that analyses videos of patients and assigns them a score based on the presence of movement patterns specific to the development of the disorder.