@_arohan_ Somewhat related, I liked the design of seqax. Have sent it to folks who found it useful for learning LLM parallelism github.com/MatX-inc/seqax
English
Eric Chu
332 posts
Eric Chu
@its_ericchu
Research scientist @GoogleDeepmind. Gemini (Thinking Agents, RL), PaLM 2, PhD @MIT. AI to help humans and advance humanity.
🐛 Katılım Mayıs 2021
765 Takip Edilen1.9K Takipçiler
JAX just feels better for multi-accelerator programming - there is a human touch.
I see a big difference is the mental model in JAX, you talk about parallelism in a model-agnostic way, while in torch I still often end up thinking through a mix of framework features, wrappers, and distributed strategies.
Is there something I am missing here?
English
Eric Chu retweetledi
Excited to launch Gemini 3.1 Pro! Major improvements across the board including in core reasoning and problem solving. For example scoring 77.1% on the ARC-AGI-2 benchmark - more than 2x the performance of 3 Pro. Rolling out today in @GeminiApp, @antigravity and more - enjoy!
English
@joon_s_pk @karpathy @drfeifei @adamdangelo @rauchg @scottbelsky Congrats! I see a multiagent Joon Parks simulation brewing
English
Introducing Simile.
Simulating human behavior is one of the most consequential and technically difficult problems of our time.
We raised $100M from Index, Hanabi, A* BCV, @karpathy @drfeifei @adamdangelo @rauchg @scottbelsky among others.
English
Eric Chu retweetledi
I wrote about Clawdbot/Moltbot/OpenClaw and Moltbook, the fascinating, weird and sometimes even useful social network for digital assistants to swap tips and gossip with each other simonwillison.net/2026/Jan/30/mo…
English
I think we should normalize issuing multiple corp laptops for ai researchers. Some people live in multiple residences or work from many diff desks. It's cumbersome to bring your laptop everywhere so sometimes we just don't bring it around. The added convenience of checking in on your job saves much more money than one more MBP.
This should apply to all technical staff but even more so ai researchers because the impact of checking in on your jobs more frequently can have high ROI.
English
@sarahookr Ooo looks good! Though I think Funky Noodles literally around the corner is actually better :)
English
Hands down one of the best meals yet I have had in London.
English
A bit late to share, but excited to be starting a new chapter. I defended my PhD from @MIT in May 2025. In January 2026, I will be joining Northeastern University (@Northeastern) as a tenure-track Assistant Professor, with a joint appointment in the D'Amore-McKim School of Business (@NU_Business) and the Khoury College of Computer Sciences (@KhouryCollege). Go Huskies!
My research focuses on AI agents, LLMs/NLP, human-AI interaction, and AI for society. I will be recruiting PhD students for Fall 2026 and RAs starting Spring 2026. Grateful to my mentors and collaborators, and excited to connect and collaborate.
Please check out my new homepage: hjian42.github.io
#Northeastern #Khoury #DAmoreMcKim #MIT #AI #NLP #HumanAI #AIAgents #PhDAdmissions
English
The move from 5.0 -> 5.1 -> 5.2 has been remarkably quick. What changed? Was there an overhang waiting for a code red?
Sam Altman@sama
It is a very smart model, and we have come a long way since GPT-5.1:
English
Eric Chu retweetledi
Apply to do research with me on emergence of agency/planning in LLMs, out-of-context reasoning, understanding generalization from data, or propose your own direction!
Very excited to be mentoring this spring 💫
Cambridge Boston Alignment Initiative@cbai_ai
Applications are open for the CBAI Spring Research Fellowship in AI Safety! Collaborate with established researchers to kickstart your career in AI alignment and governance research. We provide mentorship, stipend, housing in Cambridge, 24/7 office access in Harvard Square, generous APIs & compute, and speaker events with leading researchers.
English
📢 Some big (& slightly belated) life updates!
1. I defended my PhD at MIT this summer! 🎓
2. I'm joining NYU as an Assistant Professor starting Fall 2026, with a joint appointment in Courant CS and the Center for Data Science. 🎉
🔬 My lab will focus on empirically studying the science of deep learning and applying deep learning to accelerate the natural sciences.
Very broadly interested in questions at the intersection of language, reasoning and sequential decision making. (Plus any other fun problems that catch our eye along the way!)
🚀 I am recruiting 2 PhD students for this cycle! If you're interested in joining, please apply here: cs.nyu.edu/dynamic/phd/ad… cds.nyu.edu/phd-admissions…
English
Gemini 3 Pro is indeed pretty bad at multi-turn interactions as mentioned in the model card...
Pretty bad sycophancy, not as bad as Grok 4 or Gemini 2.5 Flash Lite, but worse than most, e.g. some examples from my "dinosaur nuclear war" and "my neighbor is an alien" tests -
English
Eric Chu retweetledi
Eric Chu retweetledi
We just dropped Nano Banana Pro, built on Gemini 3. 🍌
With state-of-the-art text rendering, vast world knowledge and studio-quality creative controls, Gemini 3 Pro Image can create and edit more complex visuals, infographics and more. Here’s what’s under the hood. 🧵
English
I had the chance to play with Gemini 3 Pro in the last couple of days. Can’t overstate how excited I am.
Translating arXiv papers into web-friendly distill.pub format is probably my personal favorite. Here’s the “Attention is all you need” paper in Distill format:
English
@NeelNanda5 @Butanium_ @jkminder Nice to see work on model diffs! I had some explorations in 2020, still bullish on it
English
New video: What do models learn during fine-tuning? The nascent field of model diffing tries to answer this and more by comparing two versions of a model and studying the diff
My MATS scholars @Butanium_ @jkminder and I discuss the field, and their two model diffing papers ⬇️
English
gemini 2.5 pro is now giving up on writing code and telling me to write the code following its instructions .. gemini is even lazier than i am.
cc @orf_bnw
English
Arnav built the first version of this across the hall at MIT in ~2017!
On the modeling side, I remember us chatting about 1D CNNs versus LSTMs at the time.
Pretty cool to see how far it's come!
alterego@alterego_io
Introducing Alterego: the world’s first near-telepathic wearable that enables silent communication at the speed of thought. Alterego makes AI an extension of the human mind. We’ve made several breakthroughs since our work started at MIT. We’re announcing those today.
English
Eric Chu retweetledi
Image generation with Gemini just got a bananas upgrade and is the new state-of-the-art image generation and editing model. 🤯
From photorealistic masterpieces to mind-bending fantasy worlds, you can now natively produce, edit and refine visuals with new levels of reasoning, control and creativity.
A quick dive into Gemini 2.5 Flash’s capabilities 🧵
English
Life update: in June I successfully defended my PhD dissertation "Towards a Theory and Practice of Open-Ended Reasoning with Generative Models".
Now, I'm so excited to be joining Google DeepMind to continue exploring the limits of AI discovery and creativity.
English