Carlos Guestrin

@guestrin

@Stanford Prof. National Acad of Eng. Chief Sci @ Visual Layer & Virtue AI. Frm Sr Dir AI @Apple. Co-author of XGBoost, LIME, TextGrad, Alpaca, TVM, GraphLab.

Stanford guestrin.stanford.edu Joined November 2009

15 Following 8.9K Followers 195 Posts

Latest posts 5

Carlos Guestrin @guestrin Oct 30

THE JOIN is the largest open collection of databases to train foundation models! We also show that it enables very effective pretraining and zero-shot generalization with RT-J. There is now a path to scale relational transformers!

rishabh ranjan Verified account @_rishabhranjan_ Oct 29

Introducing 💥RT-J💥, a System 1 model for Data
Science: put raw data in context, predict in one forward pass⚡No feature engineering, no model training, just accurate predictions.

Excited to release models, pretraining data+code and NeurIPS paper under the Stanford Tabular and Relational (STAR) ecosystem (links in comments).

How did we build this? 👇🧵

View quoted post on X

Carlos Guestrin @guestrin Jan 29

With SDPO, you can now do RL with natural language feedback, like error messages from coding environments or LLMs as judges. You can achieve huge gains over GRPO with scalar rewards!

Jonas Hübotter Verified account @jonashubotter Jan 29

Training LLMs with verifiable rewards uses 1bit signal per generated response. This hides why the model failed.

Today, we introduce a simple algorithm that enables the model to learn from any rich feedback!
And then turns it into dense supervision.

(1/n)

View quoted post on X

Carlos Guestrin @guestrin Jan 27

By specializing the model to each hard task using a simple entropic objective and RL, TTT-Discover writes new high-performing GPU kernels and improves open mathematical bounds.

Mert Yuksekgonul Verified account @mertyuksekgonul Jan 22

How to get AI to make discoveries on open scientific problems?

Most methods just improve the prompt with more attempts. But the AI itself doesn't improve.

With test-time training, AI can continue to learn on the problem it’s trying to solve: test-time-training.github.io/discover.pdf

View quoted post on X

Carlos Guestrin @guestrin Jan 22

Relational Transformers provide zero shot predictions for complex databases!

rishabh ranjan Verified account @_rishabhranjan_ Oct 30, 2025

Transformers are great for sequences, but most business-critical predictions (e.g. product sales, customer churn, ad CTR, in-hospital mortality) rely on highly-structured relational data where signal is scattered across rows, columns, linked tables and time.
Excited to finally share what I have been working on over the last year: a Foundation Model architecture which brings the power of Transformers to relational domains, enabling large-scale pretraining and zero-shot generalization in enterprise settings. 🧵1/n

View quoted post on X

Carlos Guestrin @guestrin Jan 13

TTT-E2E compresses the context simply through end-to-end test-time training — outperforms the loss of transformers with full attention, and achieves constant-time latency per token.

Karan Dalal Verified account @karansdalal Jan 12

LLM memory is considered one of the hardest problems in AI.

All we have today are endless hacks and workarounds. But the root solution has always been right in front of us.

Next-token prediction is already an effective compressor. We don’t need a radical new architecture. The missing piece is to continue training the model at test-time, using context as training data.

Our full release of End-to-End Test-Time Training (TTT-E2E) with @NVIDIAAI, @AsteraInstitute, and @StanfordAILab is now available.

Blog: nvda.ws/4syfyMN
Arxiv: arxiv.org/abs/2512.23675

This has been over a year in the making with @arnuvtandon and an incredible team.

View quoted post on X