🎉 Honored that some of our papers are ranked in the Top Articles (2020–2024) by Google Scholar Metrics 2025: 🏆 Rank #1 — Kvasir-SEG @ MMM 🏆 Rank #9 & #10 — @ IEEE JBHI 🏆 Rank #16 — HyperKvasir @ Scientific Data 🏆 Ranks #34 & #43 — @ MICCAI 🏆 Rank #77 — @ IEEE TNNLS

🆕 Releasing our entire RL + Reasoning track! featuring: • @willccbb, Prime Intellect • @GregKamradt, Arc Prize • @natolambert, AI2/Interconnects • @corbtt, OpenPipe • @achowdhery, Reflection • @ryanmart3n, Bespoke • @ChrSzegedy, Morph with special 3 hour workshop from:

Reinforcement Learning of Large Language Models, Spring 2025(UCLA) Great set of new lectures on reinforcement learning of LLMs. Covers a wide range of topics related to RLxLLMs such as basics/foundations, test-time compute, RLHF, and RL with verifiable rewards(RLVR).

Honored to present Explainable Transformers & Mamba Models for Medical Image Interpretation at IIT Roorkee & UPES Dehradun’s FDP. ✅ Showcased real-world XAI techniques in radiology ✅ Explored challenges in building trustworthy AI ✅ Engaged 41+ participants #ExplainableAI

Our paper “Mamba Guided Boundary Prior Matters” is accepted at #MICCAI2025 We introduce SAM-MaGuP – a boundary-aware, Mamba-powered upgrade to Segment Anything for polyp segmentation 🔬 SOTA across 5 datasets. 👨‍⚕️ Better detection = better cancer prevention. #AI #MedicalImaging

Honored to be named a 2024 Top Scholar by ScholarGPS, ranking in the top 0.5% globally. #51 in Image Segmentation #1,201 in Medical Imaging Grateful for this recognition! #AI #MedicalImaging #TopScholar

Our GastroVision dataset is powering @EndoML_ai’s launch project on Barrett’s vs. Normal Esophagus. A proud moment for open, clinically relevant AI in GI endoscopy. Explore the platform (no-code AI): https://t.co/l8KdiUBdbv Dataset: https://t.co/j4jfvd7lJL

🚨 Honored to deliver 2 invited talks at AICTE-FDP (MNIT Jaipur): 🎯 Reducing Miss Rates in GI Endoscopy: A Data-Centric AI Approach 🎯 Revealing the Unseen: Deep Learning to Detect Clinically Silent Patterns in Radiology 👥 150+ participants from India & abroad

Hands-On Large Language Models! This repository contains the complete code examples from the book Hands-On Large Language Models. It includes notebook examples that cover everything from the introduction to language models to fine-tuning them. 100% Open Source

🚨 BREAKING: Microsoft just dropped an 18-episode series called "Generative AI for Beginners". Ideal for beginners, developers, and AI enthusiasts looking to build a solid foundation. Here’s a breakdown (Save this👇):🧵

Goodbye Sora, Stanford just dropped FramePack, and it's INSANE 13 Wild Examples so far (Don't miss the 5th one)

Labeling images for endoscopy AI can be fun!

📢Exciting announcement today 📢 #EndoDINO is a foundation model for #GI #Endoscopy. We believe this represents a new paradigm for #AI development in the field! EndoDINO - inspired by @AIatMeta - will democratize/accelerate new AI research in gastroenterology. More below 👇

Dear incoming Master's & PhD students: 📄 CVs are static. 📊 CGPA is just a number. 💻 GitHub is your live résumé. In 2025, real-world skills > grades. Push code, not just grades. Document your work. Share your impact :) #GitHub #PhDLife #CareerGrowth

🏆 Honored to receive the Best Paper Award from @ACM_Healthcare — 5 years after publication! This was my first PhD paper, where I developed ML models using handcrafted features, while @vajira focused on DL approaches for GI disease classification. 📄 https://t.co/gXqQRmFqVG

Less than 24 hours ago, OpenAI dropped GPT-4.1. But most people missed the secret: they dropped a new prompting guide. 10 powerful tips to unlock its full potential: 👇

I just shared a new article, "The State of Reasoning Models", where I am exploring 12 new research articles on improving the reasoning capabilities of LLMs (all published after the release of DeepSeek R1): https://t.co/ric74qdVSu 1. S1: Simple test-time scaling 2. Test-Time

OpenAI published their official GPT-4.1 prompting guide, and I summarized it into these 13 practical tips to help you get the most out of the new model.

Happy to share UKBOB, the largest and most diverse 3D medical imaging segmentation dataset. It covers 72 Organs and Bones from over 51K 3D MRI samples and ONE BILLION masks from UK Biobank. This dataset enables the training of foundation models that zero-shot generalize to

Just gave an invited talk at Stanford MedAI: "From Data to Diagnosis – Advancing Medical Imaging with AI." Showcased: • Curated datasets: CirrMRI600+, PolypDB, • Robust algorithms: ColonSegNet (used by NVIDIA), ResUNet++, DoubleUNet, SAM-Mamba 🎥 Watch: https://t.co/3DoWAPALTL