RT @yidingjiang: A mental model I find useful: all data acquisition (web scrapes, synthetic data, RL rollouts, etc.) is really an explorati….

RT @ZhengyangGeng: now the code is up here:

RT @maksym_andr: 🚨Excited to release OS-Harm! 🚨. The safety of computer use agents has been largely overlooked. We created a new safety b….

RT @_vaishnavh: Wrote my first blog post! I wanted to share a powerful yet under-recognized way to develop emotional maturity as a research….

RT @YixuanEvenXu: ✨ Did you know that NOT using all generated rollouts in GRPO can boost your reasoning LLM? Meet PODS! We down-sample roll….

RT @haok1402: Introducing FLAME-MoE: a fully open platform for Mixture-of-Experts (MoE) research. All code, data, checkpoints, training log….

RT @ZhengyangGeng: Excited to share our work with my amazing collaborators, @Goodeat258, @SimulatedAnneal, @zicokolter, and Kaiming. In a….

RT @pratyushmaini: Excited to be talking today about how research into memorization provides a fundamentally different lens on safety!.

RT @RuntianZhai: A shorter version of the first three chapters of my thesis is accepted by ICML 2025. It provides a quick start for those i….

RT @pratyushmaini: Looking forward to giving a talk this Friday @OpenAI with @zhilifeng on some of our privacy & memorization research + ho….

RT @electronickale: ✨ Love 4o-style image generation but prefer to use Midjourney? Tired of manual prompt crafting from inspo images?. PRIS….

RT @_christinabaek: When we train models to do QA, are we robustly improving context dependency? No!. In our ICLR Oral (Fri 11 AM), we show….

RT @kchonyc: spicy @zicokolter

Arxiv: Website: Talk: Also with @zhilifeng @A_v_i__S @AlexRobey23 @m_finzi.

Excited about this work with @ashertrockman @yashsavani_ (and others) on antidistillation sampling. It uses a nifty trick to efficiently generate samples that makes student models _worse_ when you train on samples. I spoke about it at Simons this past week. Links below.

Thanks @NVIDIADC for the DGX B200 machine for the CMU Catalyst group! I'm perhaps already a bit too enthralled by it in the photos. .

RT @SCSatCMU: Huge thank you to @NVIDIADC for gifting a brand new #NVIDIADGX B200 to CMU’s Catalyst Research Group! This AI supercomputing….

RT @_christinabaek: Are current reasoning models optimal for test-time scaling? 🌠.No! Models make the same incorrect guess over and over ag….

RT @pratyushmaini: 1/Being in academia is such a privilege: You get to collaborate with insanely talented & passionate students on their jo….

RT @FahimTajwar10: Interacting with the external world and reacting based on outcomes are crucial capabilities of agentic systems, but exis….