Armielyn Obinguar
@Aeriumcius
developer relations @tarsprotocol| vibe ++ generative ai with research papers and products | prev @virtuals_io
Bullish for $TAI way to go! @tarsprotocol We cookin’
TARS AI was on the ground at Google Cloud Summit Nord in Düsseldorf, surrounded by builders, AI engineers, and cloud leaders. From autonomous agents to GenAI infrastructure, the takeaway was clear, the market is moving from pilots to large-scale deployment. For TARS AI, it was…
Day 2 of #MYB2025 The view is so amazing here in Exchange 106!
My Reinforcement Learning (RL) & Agents 3 hour workshop is out! I talk about: 1. RL fundamentals & hacks 2. "Luck is all you need" 3. Building smart agents with RL 4. Closed vs Open-source 5. Dynamic 1bit GGUFs & RL in @UnslothAI 6. The Future of Training youtube.com/watch?v=OkEGJ5…
#MYB2025 hosting the institutional night here🫶 Been trying the local foods so far the Nasi Lemak is my fav!
Here at #MYBW2025 proudly representing @tarsprotocol sharing what we’ve been building at the intersection of Web3 and tech. Great to connect with curious minds, passionate builders, and new collaborators across the space. If you’re around, let’s connect!
"experts" for harder tokens? "Mixture-of-Recursions (MoR): Learning Dynamic Recursive Depths for Adaptive Token-Level Computation" MoR makes one shared Transformer block loop only for tokens that need extra thought, delivering quality with half the weights & twice the speed
We're thrilled to announce our new course: Retrieval Augmented Generation (RAG) RAG is a key part of building LLM applications that are grounded, accurate, and adaptable. In this course, taught by AI engineer @ZainHasan6 and available on @Coursera, you’ll learn how to design…
This is my lecture from 2 months ago at @Cornell “How do I increase my output?” One natural answer is "I will just work a few more hours." Working longer can help, but eventually you hit a physical limit. A better question is, “How do I increase my output without increasing…
Flow Q-learning (FQL) is a simple method to train/fine-tune an expressive flow policy with RL. Come visit our poster at 4:30p-7p this Wed (evening session, 2nd day)!
Excited to introduce flow Q-learning (FQL)! Flow Q-learning is a *simple* and scalable data-driven RL method that trains an expressive policy with flow matching. Paper: arxiv.org/abs/2502.02538 Project page: seohong.me/projects/fql/ Thread ↓
Authored by 3000+ researchers, the Google Gemini Team has published a technical report on the Gemini 2.X family They utilized sparse MoE with a "Thinking" module which enables internal reasoning before generating Their heavy focus on long ctx made agentic app incredibly good
Unfortunately, I won't be able to attend #ICML2025 in person due to visa delays. But I'm excited to share our paper in ICML: "Large Language Models are Demonstration Pre-Selectors for Themselves"! 🧠📄 💡 What if LLMs could help pick better examples for themselves? We propose…
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).
In case you missed it, we recently launched "Post-training of LLMs," a short course where you'll: ✅ Understand when and why to use post-training methods like Supervised Fine-Tuning (SFT), Direct Preference Optimization (DPO), and Online Reinforcement Learning. ✅ Learn the…