🔥The 9th Round of Easy Loan, Earn $40 Reward is in progress❗️ ⏰ Promotion Period: January 15th - Feburary 15th, 2025 👉 Register now and check more details at gate.io/campaigns/358

# RLHF is just barely RL Reinforcement Learning from Human Feedback (RLHF) is the third (and last) major stage of training an LLM, after pretraining and supervised finetuning (SFT). My rant on RLHF is that it is just barely RL, in a way that I think is not too widely

This is the most gut-wrenching blog I've read, because it's so real and so close to heart. The author is no longer with us. I'm in tears. AI is not supposed to be 200B weights of stress and pain. It used to be a place of coffee-infused eureka moments, of exciting late-night arxiv

MolLEO is accepted ICLR 2026! We have made so much progress to show LLMs really have tons of knowledge about science and it’s not just retrieving. LLMs easily beat SOTA molecule optimization methods with an evolutionary process!

We're launching new tools to help developers build reliable and powerful AI agents. 🤖🔧 Timestamps: 01:54 Web search 02:41 File search 03:22 Computer use 04:07 Responses API 10:17 Agents SDK

🧠 Excited to share our latest work: "MedAgentsBench: Benchmarking Thinking Models and Agent Frameworks for Complex Medical Reasoning"! We've curated a challenging hard subset from existing medical QA datasets. We select questions where fewer than 50% of the LLMs (incl. GPT-4o,

Today, we're excited to make two big announcements! - OpenHands LM: The strongest 32B coding agent model, resolving 37.4% of issues on SWE-bench Verified 📈 - OpenHands Cloud: SOTA open-source coding agents from your computer, phone, github, with $50 in free credits 🙌☁️

We’re releasing PaperBench, a benchmark evaluating the ability of AI agents to replicate state-of-the-art AI research, as part of our Preparedness Framework. Agents must replicate top ICML 2024 papers, including understanding the paper, writing code, and executing experiments.

This is so cool!

MLE-Dojo: Interactive Environments for Empowering LLM Agents in Machine Learning Engineering A Gym-style framework for systematically training, evaluating, and improving agents in iterative ML engineering workflows

Georgia Tech and Stanford Researchers Introduce MLE-Dojo: A Gym-Style Framework Designed for Training, Evaluating, and Benchmarking Autonomous Machine Learning Engineering (MLE) Agents Researchers from Georgia Institute of Technology and Stanford University have introduced

We’ve developed Gemini Diffusion: our state-of-the-art text diffusion model. Instead of predicting text directly, it learns to generate outputs by refining noise, step-by-step. This helps it excel at coding and math, where it can iterate over solutions quickly. #GoogleIO

🤔 How can we systematically enhance LLMs for complex medical coding tasks? 🚀 Introducing MedAgentGym, an interactive gym-style platform designed specifically for training LLM agents in coding-based medical reasoning! 🧬💻 🎯 Comprehensive Code-based Medical Reasoning

Excited to share that I joined Google DeepMind as a research scientist recently 🥳 Looking forward to future collaborations on exciting projects :)

This Monday, Wenqi Shi from UT Southwestern Medical Center will be joining us to talk about their work on training LLM agents for code-based medical reasoning at scale. Catch it at 1-2pm PT this Monday on Zoom! Subscribe to mailman.stanford.edu/mailman/listin… #ML #AI #medicine #healthcare

MedAgentGym is the first publicly available training environment designed to improve the ability of LLMs to use code for medical reasoning tasks. It includes 72,413 instances from 12 benchmarks. Tasks are run in isolated, executable environments that provide interactive feedback.

Gemini + Deep Think won IMO gold this year 🏅 super honored to be part of this dream team!

Presenting Zebra-CoT: A large-scale dataset to teach models intrinsic multimodal reasoning: interleaving text and natively-generated images like a zebra's stripes. It moves beyond the limitations of external tool-based visual CoT. 🔗arxiv.org/abs/2507.16746

Gemini 2.5 Deep Think is available to Ultra users! It achieves SOTA on HLE (no tools), LiveCodeBench, and math/proofs. Time to give it a try and let us know your feedback! We’ve also made the IMO gold model available to mathematicians and other domain experts :)👩‍🍳

Gemini 2.5 Deep Think is available to Ultra users!🥳