Vaishnavh Nagarajan

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

I liked a few from recent Simons Institute Workshops: youtube.com/live/9V0bfZqT1… youtube.com/live/M7Kq0ooFF… youtube.com/live/nplao1_t8…

A similar study for diffusion has also been reported here: arxiv.org/html/2411.1933… Similarly, they show that diffusion fails to estimate the ¨optimal¨ score

I’m presenting two papers on value-based RL for post-training & reasoning on Friday at @ai4mathworkshop at #ICML2025! 1️⃣ Q#: lays theoretical foundations for value-based RL for post-training LMs; 2️⃣ VGS: practical value-guided search scaled up for long CoT reasoning. 🧵👇

1/So much of privacy research is designing post-hoc methods to make models mem. free. It’s time we turn that around with architectural changes. Excited to add Memorization Sinks to the transformer architecture this #ICML2025 to isolate memorization during LLM training🧵

I will be at #ICML2025 🇨🇦 from Wednesday through Saturday. My students have a lot of exciting papers - check them out and come talk to us! Especially thrilled to have received the Outstanding Paper Award🏆 for our work on creativity.

Today at #ICML2025, we present Deliberate Practice: an approach to improve sample-efficiency by generating harder, not more, examples. - Oral talk at 10:45 - West Ballroom B | Orals 3C: Data-Centric ML Join us to discuss principled approaches to more efficient learning.

Today @ChenHenryWu and I will be presenting our #ICML work on creativity in the Oral 3A Reasoning session (West Exhibition Hall C) 10 - 11 am PT Or please stop by our poster right after @ East Exhibition Hall A-B #E-2505 11am-1:30pm. (Hope you enjoy some silly human drawings!)

thank you! tagging @ChenHenryWu, the student lead on this, @chuckzduck (that's Charles btw) and @AdtRaghunathan! also tagging @GregorBachmann1 who led the previous work we built on

Very excited for a new #ICML2025 position paper accepted as oral w @mbodhisattwa & @TuhinChakr! 😎 What are the longitudinal harms of AI development? We use economic theories to highlight AI’s intertemporal impacts on livelihoods & its role in deepening labor-market inequality.

I'm at ICML presenting this work! Come by on Tuesday to hear about/chat about combinatorial generalization and creativity in diffusion models. x.com/MasonKamb/stat…

We have exactly the same claim in our natural language RL arxiv.org/abs/2411.14251. I cannot believe even the phrases are so similar. I believe this is what differs traditional RL and learning from experience -- we should learn more from experience, not just reward.