🔥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

We show some pretty intriguing results with massive implications! Tune into Aditi’s talk at SCSL workshop, 1:30pm on Monday, Garnet 214-215 (Floor 2). #ICLR2025

Building LLM Agents? Come to my talk at the #ICLR DATA-FM workshop today at 2:30pm, Hall 4, Section 4. I'll be presenting InSTA, our work building the largest environment for agents on the live internet. arxiv.org/abs/2502.06776 #Agents #LLM

✨ Love 4o-style image generation but prefer to use Midjourney? Tired of manual prompt crafting from inspo images? PRISM to the rescue! 🖼️→📝→🖼️ We automate black-box prompt engineering—no training, no embeddings, just accurate, readable prompts from your inspo images! 1/🧵

Our paper on how overtraining LLMs can make fine-tuning harder won awards at two different #ICLR2025 workshops! I'm honored and thrilled! Outstanding paper @ SCOPE Entropic Paper Award @ ICBINB

Happy to share that Compositional Risk Minimization has been accepted at #ICML2025 📌Extensive theoretical analysis along with a practical approach for extrapolating classifiers to novel compositions! 📜 arxiv.org/abs/2410.06303

Looking forward to giving a talk this Friday OpenAI with Zhili Feng on some of our privacy & memorization research + how it applies to production LLMs! We've been gaining momentum on detecting, quantifying & erasing memorization; excited to explore its real-world impact!

Winner of the Entropic Paper Award at ICLR 2026 #ICLR2025 Groundbreaking research by Jacob Mitchell Springer Jacob Springer (CMU), Sachin Goyal Sachin Goyal (CMU), Kaiyue Wen Kaiyue Wen (Stanford), Tanishq Kumar (Harvard), Xiang Yue Xiang Yue (CMU), Sadhika Malladi

"RL with only one training example" and "Test-Time RL" are two recent papers that I found fascinating. In the "One Training example" paper the authors find one question and ask the model to solve it again and again. Every time, the model tries 8 times (the Group in GRPO), and

Excited to share our work with my amazing collaborators, Goodeat, Xingjian Bai, Zico Kolter, and Kaiming. In a word, we show an “identity learning” approach for generative modeling, by relating the instantaneous/average velocity in an identity. The resulting model,

Very interesting work!

follow Aditi Raghunathan for all the other exciting LLM stuff going on in Aditi's group! i owe a lot to the group & its students. :) they have the most liveliest slack channel I've been on, keeping me up to date with all the latest AI stuff, great ideas & thoughtful discussions

So true!

New model drop - Sarvam-Translate is here. Can translate between 22 Indian languages & English. Significantly better than much larger models. Improves on nuance, long-form, structured text. Available as super-fast APIs. Try it here: dashboard.sarvam.ai/translate

Excited to speak at the CVPR workshop on domain generalization! Estimating model performance in the wild is hard but crucial. I'll present agreement-on-the-line, a simple and surprisingly powerful phenomenon. It is easily one of the most intriguing things I've studied.

Wrote my first blog post! I wanted to share a powerful yet under-recognized way to develop emotional maturity as a researcher: making it a habit to read about the ✨past ✨ and learn from it to make sense of the present

📢We show that continuous latent reasoning has a theoretical advantage over discrete token reasoning (arxiv.org/abs/2505.12514): For a graph with n vertices and graph diameter D, a two-layer transformer with D steps of continuous CoTs can solve the directed graph reachability

Pressing reload on 10's of VS Code windows in the morning is the adult version of reload in Counter-Strike 🔃.

All set for the #ICML grind after the #GrouseGrind⛰️ Eager to discuss my recent works on quantifying, detecting, & eliminating model memorization. I'll be at one of the events below when not at the DatologyAI booth. Pls DM if you'd like to chat about data quality or privacy!

Presenting CRM at #ICML2025 📌 Wednesday, 16th July, 11 am 📍East Exhibition Hall A-B (E-2101) Lets chat about distribution shifts! Been deep into causality & invariance based perspectives, and recently exploring robust LLM pretraining architectures.

Some updates 🚨 I finished my Ph.D at Allen School in June 2025! After a year at AI2 as a Research Scientist, I am joining CMU Language Technologies Institute | @CarnegieMellon & Machine Learning Dept. at Carnegie Mellon (courtesy) as an Assistant Professor in Fall 2026. The journey, acknowledgments & recruiting in 🧵