Загрузка видео...
Не удалось загрузить видео
🚨 Just released: MIRIAD, a million-scale medical QA dataset to ground LLMs in reliable medical knowledge. 5.8M question-answer pairs, each distilled from peer-reviewed literature! 🔥 That's structured, high-quality data built for medical AI. 🧵 ↓
48,127 просмотров • 1 год назад •via X (Twitter)
Комментарии: 12
Why this matters: Most RAGs or agentic systems rely on raw text: messy, noisy, and poorly aligned with downstream tasks. MIRIAD offers an alternative: ↳ Curated QA pairs ↳ Structured for retrieval ↳ Optimized for grounding and evaluation
What you can do with MIRIAD: - Boost RAG/agentic system performance - Train medical retrievers - Detect hallucinations in clinical LLMs - Build knowledge-grounded expert assistants
Also included: MIRIAD Atlas, an interactive map of 56 specialties that lets you explore, search, and trace answers back to the original source literature. Semantic search built in! 🔥
The dataset was built through a semi-automated pipeline: LLM rephrasing → automatic filtering → multi-level QC → expert human annotation
🤗 Dataset on @huggingface:
🔗 Website:
📜 Preprint:
💻 Code:
🪐 Demo:
Credits for this great work! Supervisor: @Michael_D_Moor Co authors: Qinyue Zheng @qinzytech Salman Abdullah @salmanabdullah_ Sam Rawal @samarthrawal Cyril Zakka, MD @cyrilzakka Sophie Ostmeier @SophieOstmeier Eric Topol @EricTopol Maximilian Purk Eduardo P. Reis Jure Leskovec
If you found this helpful, a like or RT goes a long way for this open source project to be discovered and used by many! Follow me → @datachaz for insights on lesser known projects like this and more content on LLMs, AI agents, and data science 🦾
🌱 The gut-lung axis may hold the key to managing asthma, COPD, and ARDS. Explore how gut health may influence respiratory diseases and the future of targeted therapies. Check out the latest research at #GutLungAxis #Microbiome #LungHealth #health #news
![🔥 Leslie Manookian: Covid woke everyone up, now it’s time to take action. We just launched a medical freedom revolution. This year, we’re bringing legislation to ban medical mandates to as many states as possible. And Leslie already delivered our biggest win ever in 2025. 99% of people didn’t notice. But everyone is about to hear from us this year. “From the Covid policies … the Overton Window actually shifted with respect to vaccines.” “People … all across our nation were waking up to the reality.” “The vaccines were not like a sip of water.” “They had a serious dark side to them.” “I thought … it’s time right now, while this is fresh in their minds, to launch a Medical Freedom Act.” “So I wrote the Medical Freedom Act… and it unbelievably passed in [Idaho].” “The Medical Freedom Coalition is an opportunity to bring as many of the major groups that we can together to flex our collective muscle.” “[So we can] get Medical Freedom Acts passed into as many states as possible this legislative session.” Find out everything you need to know about our new Medical Freedom Act Coalition—and how to support us—below:🧵 Leslie Manookian Feds For Freedom](https://image.24vids.com/tw-2018375617801080979/amplify_video_thumb/2018375504823246848/img/4jUKIIYAVWn4sEkT.jpg)
![Studies have shown ChatGPT outperforms human annotators for Structured Data by about 25% and costs 30x less. 1 In just 2 months, miners on SN33 running ChatGPT without optimization can’t survive. Today we announce SN33 is now ReadyAI to fully align with our mission 👇 SN33 is building a more performant and significantly cheaper alternative to Scale AI Today structured data is performed primarily by human annotation services like Amazon’s Mechanical Turk and Scale AI It is now more important than ever for every business and individual to make their data AI Ready. However, taking unstructured data and making it Structured Data using today’s tools is extremely costly. SN33 revolutionizes this process, unlocking immense opportunities for commercialization. We lay out the vision for it in this detailed blog post: Validators TODAY can monetize access to this structured data pipeline independently, but we’re streamlining this process, launching a frontend soon that any validator can opt into to provide bandwidth. We've received great feedback from the community, recognizing that what we're building goes far beyond Conversational AI. Building the world's largest annotated conversational dataset (which we've already accomplished) is just one of countless real-world applications for SN33's Structured Data pipeline. We're building a decentralized Scale AI, offering a full suite of Structured Data commodities—from text metadata tagging (available today) to fully customizable queries for company-specific data annotation use cases and image metadata tagging coming soon 👀. Thanks for all the feedback! It has been invaluable so keep bringing it to us! 🙏$TAO Openτensor Foundaτion 1 “ChatGPT Outperforms Crowd-Workers for Text-Annotation Tasks” shows “The zero-shot accuracy of ChatGPT exceeds that of crowd-workers by about 25 percentage points on average [...] Moreover, the per-annotation cost of ChatGPT is less than $0.003—about thirty times cheaper than MTurk”](https://image.24vids.com/tw-1825631402781819359/ext_tw_video_thumb/1825631116633714689/pu/img/ihH0N-4TGHxyNXtb.jpg)
