Rui Yang

🤖Can MLLM agents reason about spatial relationships and plan atomic actions for navigation & manipulation? 🔥 Meet EmbodiedBench 🏆—the first fine-grained benchmark for MLLM-based embodied agents! 📄 Paper: arxiv.org/abs/2502.09560 🌐 Website & code: embodiedbench.github.io

🔥Our Goedel-Prover-V2-32B topped the PutnamBench Leaderboard by solving 86 problems —nearly 2× more than the previous SOTA DeepSeek-Prover-V2-671B (solved 47), while using: * 1/20 the model size (32B vs. 671B) * 1/5 the passes (184 vs. 1024) Meanwhile, we also release *…

🚀 RL is powering breakthroughs in LLM alignment, reasoning, and agentic apps. Are you ready to dive into the RL x LLM frontier? Join us at @aclmeeting ACL’25 tutorial: Inverse RL Meets LLM Alignment this Sunday at Vienna🇦🇹(Jul 27th, 9am) 📄 Preprint at huggingface.co/papers/2507.13…

Grateful for the chance to present EmbodiedBench at ICML as an Oral. A rewarding experience full of learning. Thanks for @RuiYang70669025 @hengjinlp @jyzhang1208 @huan_zhang12 Mark_Zhao @ManlingLi_ Tong_Zhang and many others who make it possible. See you next time.

🚀 Introducing MA-LoT Theorem Framework: An open-source multi-agent framework utilizing the Long Chain-of-Thought to boost automated theorem-proving🎉 ✅ Achieving 61.07% accuracy rate under pass@32 on MiniF2F-Test outperforming Goedel-Prover, Lean_STP and DeepSeek-Prover-V1.5…

(1/4)🚨 Introducing Goedel-Prover V2 🚨 🔥🔥🔥 The strongest open-source theorem prover to date. 🥇 #1 on PutnamBench: Solves 64 problems—with far less compute. 🧠 New SOTA on MiniF2F: * 32B model hits 90.4% at Pass@32, beating DeepSeek-Prover-V2-671B’s 82.4%. * 8B > 671B: Our 8B…

Learning to perceive while learning to reason! We introduce PAPO: Perception-Aware Policy Optimization, a direct upgrade to GRPO for multimodal reasoning. PAPO relies on internal supervision signals. No extra annotations, reward models, or teacher models needed. 🧵1/3

🤩Mind-blowing discovery: Random policies can be surprisingly powerful for decision-making! Our ICML 2025 paper reveals how simple randomness leads to sophisticated reward-matching policies. Let me break this down...

Awesome work! 🥂 I feel like the design of our GUI-Actor — which can propose multiple candidate regions in one forward pass— combined with a Grounding Verifier could work really well within the 'test-time scaling' framework of GTA1! 😀

Insightful post on the scalability of off-policy RL.

🧵 1/7 Should AI agents "think more" or "do more"? 🤔 The current trend is to scale test-time compute, making agents generate longer reasoning traces. But what if that’s the wrong approach for interactive tasks? In our new work, we argue for a new scaling dimension: Test-Time…

🚀 Can LLMs stop overthinking when detailed reasoning isn't needed? Excited to share our latest work on LLM reasoning: AutoL2S 🧠⚡ 📄 Paper: arxiv.org/abs/2505.22662 🤖 Model: huggingface.co/amandaa/AutoL2… LLMs often overthink—generating unnecessarily long CoTs even for easy…

Excited to share that EmbodiedBench was selected for an Oral at ICML 2025! We recently added results for new models (InternVL3, Gemma3, Ovis2) and released a large agent trajectory dataset on 🤗: embodiedbench.github.io Try training and evaluating your MLLM for embodied agents!

🚀 Excited to share GUI-Actor—a new approach for GUI grounding! Big thanks to @_akhaliq for featuring our work! 🌐 Project page: microsoft.github.io/GUI-Actor/ 📜 Paper: arxiv.org/pdf/2506.03143 🤔 What's limiting coordinate generation-based GUI grounding? 1️⃣ Weak spatial-semantic…

Thanks for sharing our work！GUI-Actor is a new GUI grounding method that combines an attention-based action head with a grounding verifier, different from previous text-based coordinate prediction methods.

Does RL truly expand a model’s reasoning🧠capabilities? Contrary to recent claims, the answer is yes—if you push RL training long enough! Introducing ProRL 😎, a novel training recipe that scales RL to >2k steps, empowering the world’s leading 1.5B reasoning model💥and offering…

📢 New Paper Drop: From Solving to Modeling! LLMs can solve math problems — but can they model the real world? 🌍 📄 arXiv: arxiv.org/pdf/2505.15068 💻 Code: github.com/qiancheng0/Mod… Introducing ModelingAgent, a breakthrough system for real-world mathematical modeling with LLMs.

Is there anything that Qwen cannot do at this point? 😂

🚀 A unified strategy for parallel decoding: Fractured CoT Reasoning We explore three dims of sampling: - Reasoning trajectories - Final solutions per traj - Depth of reasoning Maximize accuracy-cost trade-off! Allocate computation for huge gains. Paper: arxiv.org/pdf/2505.12992