Querri’s Post

We recently launched our #AI assistant, YOKOTAI, for automating data-driven workflows with the help of language models. One of the most frequent questions we get is about the data types that can be uploaded. There is a lot of interest around tabular data like spreadsheets especially, and whether #LLMs can understand it. Yes, LLMs can work with tabular data! But they shouldn't work directly on the raw data - instead we need to give them access to proper tools. #PandasAI is one such tool. Here's an example of what we can do. I've downloaded data from OpenLLM benchmarks hosted by #Huggingface, showing performance of openly available language models. With GPT-4, I can simply ask questions like which model is the best for zero shot tasks and how much better is it than the rest. Or I can ask it to draw plots from the spreadsheet! The picture shows the answers and the plot we generated. So yeah, tabular data can be understood by LLMs :) Code in comments!

#AI is impressive, but it's not perfect. When it comes to #SQL, trusting AI without human expertise is risky. ❌ - Inaccurate code & bad decisions - Lack of understanding = misleading answers - Importance of expert review AI can be a powerful tool, but it should empower data experts, not replace them. Read more about the risks & best practices for AI-Generated SQL: https://bit.ly/3T0FKil

🌐🤖 Exciting News! Introducing InfiniQuery: The Infinite Context QA Companion! 📜📄 Have you ever wished for an AI system that could provide context-aware answers based on the latest research papers? Look no further! InfiniQuery is here to revolutionize the way we interact with information. 🔍 What is InfiniQuery? InfiniQuery is an advanced AI contextual question-answering system built on the groundbreaking “Leave No Context Behind” research paper. It combines the power of Large Language Models (LLMs) like Gemini 1.5 Pro with external data sources (PDFs, text files, etc.) to provide accurate and up-to-date answers. 🌟 Key Features: Infinite Context: InfiniQuery can handle extensive context, making it ideal for complex questions that require deep understanding. Retrieval-Augmented Generation (RAG): Our system seamlessly integrates retrieval and generation models, ensuring precise and context-aware responses. 🚀 How It Works: Retrieve: InfiniQuery fetches relevant information from the “Leave No Context Behind” paper. Generate: It passes this context to Gemini 1.5 Pro, which generates informed responses. Answer: Get accurate answers based on the paper’s content! 👩💻 Use Cases: Researchers: Quickly access relevant details from research papers. Students: Enhance your understanding of complex topics. Professionals: Stay informed about industry trends. Here is the link: https://lnkd.in/g9DhphBr Special Thanks to Kanav Bansal sir, Raghu Ram Aduri sir and Innomatics Research Labs team for their support! #AI #ContextualQA #Research #Innovation #InfiniQuery #InnomaticsResearchLabs 📢 Let’s leave no context behind! Share this post and spread the word about InfiniQuery! 🚀🔥

"A look at the ‘data plumbing’ that powers AI and companies like Grab, Coins.ph, and Maya" Rappler in X by Lance Spencer Yu A #unifieddataplatform is super important. Some of these basic data works may not sound very sexy but is the #foundation if you want to do any of the sexy stuff. If you don’t have all your data in one unified place in one format that is accessible by all the #dataengineers, #dataanalysts, and #datascientists then you end up with siloed #data pieces - Ng told reporters. - Cecily Ng #genAI #generativeAI https://lnkd.in/dReSgitB

Lots of AI techniques to cover in this second whisky & analytics post - check it out with the link below! I wanted to dive into the heart of the modern #AI stack: covering large language models, storage of word embeddings with vector databases, retrieval-augmented generation (#RAG) to bring together the power of #LLMs and my whisky reviews data, and #LangChain as the glue for my AI application! For fun, I also added some cool data visualization using #Spotlight from Renumics to explore the nearest neighbours in my embeddings space - what whisky gems did I uncover?! Slàinte mhath! https://lnkd.in/erMWke98

Autonomous Database now speaks human! The new Select AI feature combines generative #AI with Oracle SQL, so users can ask questions in natural language to get answers from their data without knowing SQL. Find out how this groundbreaking capability works: https://lnkd.in/eXsmXpdN

Autonomous Database now speaks human! The new Select AI feature combines generative #AI with Oracle SQL, so users can ask questions in natural language to get answers from their data without knowing SQL. Find out how this groundbreaking capability works: https://lnkd.in/ewkSsaW8