🔥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

😄I did a brief intro of RLHF algorithms for the reading group presentation of our lab. It was a good learning experience for me and I want to share the github repo here holds the slides as well as the list of interesting papers: github.com/yihedeng9/rlhf… Would love to hear about

🚀 Excited to share our position paper: "Formal Mathematical Reasoning: A New Frontier in AI"! 🔗 arxiv.org/abs/2412.16075 LLMs like o1 & o3 have tackled hard math problems by scaling test-time compute. What's next for AI4Math? We advocate for formal mathematical reasoning,

Cool things from DeepSeek v3's paper: 1. Float8 uses E4M3 for forward & backward - no E5M2 2. Every 4th FP8 accumulate adds to master FP32 accum 3. Latent Attention stores C cache not KV cache 4. No MoE loss balancing - dynamic biases instead More details: 1. FP8: First large

Interested in the combination of Inference time scaling + LLM Agent?🤖💭 Announcing QLASS (Q-guided Language Agent Stepwise Search, arxiv.org/abs/2502.02584), a framework that supercharges language agents at inference time. ⚡In this work, we build a process reward model to guide

New paper & model release! Excited to introduce DuoGuard: A Two-Player RL-Driven Framework for Multilingual LLM Guardrails, showcasing our new DuoGuard-0.5B model. - Model: huggingface.co/DuoGuard/DuoGu… - Paper: arxiv.org/abs/2502.05163 - GitHub: github.com/yihedeng9/DuoG… Grounded in a

Introducing #SIRIUS🌟: A self-improving multi-agent LLM framework that learns from successful interactions and refines failed trajectories, enhancing college-level reasoning and competitive negotiations. 📜Preprint: arxiv.org/pdf/2502.04780 💻code: github.com/zou-group/siri… 1/N

🚀 Exciting news! Our Goedel-Prover paper is now live on arXiv: arxiv.org/pdf/2502.07640 🎉 We're currently developing the RL version and have a stronger checkpoint than before (currently not included in the report)!🚀🚀🚀 Plus, we’ll be open-sourcing 1.64M formalized

🚀 Introducing NSA: A Hardware-Aligned and Natively Trainable Sparse Attention mechanism for ultra-fast long-context training & inference! Core components of NSA: • Dynamic hierarchical sparse strategy • Coarse-grained token compression • Fine-grained token selection 💡 With

🌟 Can better cold start strategies improve RL training for LLMs? 🤖 I’ve written a blog that delves into the challenges of fine-tuning LLMs during the cold-start phase and how the strategies applied there can significantly impact RL performance in complex reasoning tasks that

[1/n] SuperExcited to announce SuperGPQA!!! We spend more than half a year to finally make it done! SuperGPQA is a comprehensive benchmark that evaluates graduate-level knowledge and reasoning capabilities across 285 disciplines. It also provides the largest human-LLM

Excited to release PrefEval (ICLR '25 Oral), a benchmark for evaluating LLMs’ ability to infer, memorize, and adhere to user preferences in long-context conversations! ⚠️We find that cutting-edge LLMs struggle to follow user preferences—even in short contexts. This isn't just

We’re rolling out Deep Research to Plus users today! Deep Research was the biggest “Feel The AGI” moment I’ve ever had since ChatGPT. And I’m glad more people will experience their first AGI moment! The team also worked super hard to make more tools including image citations /

🤖 I just updated my repository of RL(HF) summary notes to include a growing exploration of new topics, specifically adding notes to projects related to DeepSeek R1 reasoning. Take a look: github.com/yihedeng9/rlhf… 🚀 I’m hoping these summaries are helpful, and I’d love to hear

🗞️Arxiv: arxiv.org/abs/2503.17352

Thanks AK for sharing our work!

Introducing d1🚀 — the first framework that applies reinforcement learning to improve reasoning in masked diffusion LLMs (dLLMs). Combining masked SFT with a novel form of policy gradient algorithm, d1 significantly boosts the performance of pretrained dLLMs like LLaDA.