Our latest post explores on-policy distillation, a training approach that unites the error-correcting relevance of RL with the reward density of SFT. When training it for math reasoning and as an internal chat assistant, we find that on-policy distillation can outperform other

🚀If your code agent generates a patch that passes all tests, should you trust it merge automatically? ⚠️You probably shouldn’t! “Correct” ≠ “Safe.” In our study we show that a single normal looking issue description, whether from a benign user or not, can lead code agents

🧠 How can we equip LLMs with memory that allows them to continually learn new things? In our new paper with @AIatMeta, we show how sparsely finetuning memory layers enables targeted updates for continual learning, w/ minimal interference with existing knowledge. While full

🤔 Can AI optimize the systems it runs on? 🚀 Introducing FlashInfer-Bench, a workflow that makes AI systems self-improving with agents: - Standardized signature for LLM serving kernels - Implement kernels with your preferred language - Benchmark them against real-world serving

🤔Can we train RL on LLMs with extremely stale data? 🚀Our latest study says YES! Stale data can be as informative as on-policy data, unlocking more scalable, efficient asynchronous RL for LLMs. We introduce M2PO, an off-policy RL algorithm that keeps training stable and

🔥 We introduce Multiverse, a new generative modeling framework for adaptive and lossless parallel generation. 🚀 Multiverse is the first open-source non-AR model to achieve AIME24 and AIME25 scores of 54% and 46% 🌐 Website: https://t.co/J9osByhWUf 🧵 1/n

I've written the full story of Attention Sinks — a technical deep-dive into how the mechanism was developed and how our research ended up being used in OpenAI's new OSS models. For those interested in the details: https://t.co/0EAi2KQMMx

Our open models are here. Both of them. https://t.co/9tFxefOXcg

Congrats to Rice CS' @Anshumali_ Shrivastava, who has been promoted to full professor. Shrivastava is well on his way to revolutionizing how LLMs & other deep learning models are trained & stored, using new algorithms to make AI scalable & more accessible. https://t.co/8VpFk371gp

Huge thanks to @tinytitans_icml for an amazing workshop — see you next year! Honored to receive a Best Paper Award 🏆 Let’s unlock the potential of sparsity! Next up: scaling to hundreds/thousands of rollouts? Or making powerful R1/K2-level LLMs (not just 8B 4-bit models) run

🚨 Super excited to share that our paper "GSM-Infinite: How Do Your LLMs Behave over Infinitely Increasing Context Length and Reasoning Complexity?" got accepted at ICML 2025! 🧠📏🎉 P.S. 😅 Struggling to make your long-context LLM submission stand out at NeurIPS? 🧠 Give