Ever wondered how nondeterministic GPT-4 is even with greedy decoding (T=0)? I built a website that asks GPT-4 to draw a unicorn every hour and tracks if the results stay consistent over time (spoiler alert: they don't! 🦄). Explore the findings: openaiwatch.com

Meet the Chalk-Diagrams plugin for ChatGPT! Based on Dan Oneață & Sasha Rush's lib, it lets you create vector graphics with language instructions. Try it with ChatGPT plugins: 1️⃣ Plugin store 2️⃣ Install unverified plugin 3️⃣ chalk-diagrams.com Test it out: "draw a pizza" 🍕

Excited to share that I'm joining Waterloo's Cheriton School of Computer Science as an Assistant Professor and Vector Institute as a Faculty Affiliate in Fall '24. Before that, I'm doing a postdoc at Ai2 with Yejin Choi. Immensely grateful to my PhD advisors Sasha Rush and Stuart Shieber (@[email protected]). This journey wouldn't have

Can LMs solve reasoning tasks without showing their work? "Implicit Chain of Thought Reasoning via Knowledge Distillation" teaches LMs to reason internally to solve tasks like 5×5 multiplication. Here's how we bypass human-like step-by-step reasoning bit.ly/implicitCoT 1/6

I am hiring NLP/ML PhD students at UWaterloo, home to 5 NLP professors! Apply by Dec 1 Strong consideration will be given to those who can tackle the below challenge: Can we use LM's hidden states to reason multiple problems simultaneously? ​​Retweets/shares appreciated🥰

What do people use ChatGPT for? We built WildVis, an interactive tool to visualize the embeddings of million-scale chat datasets like WildChat. Work done with Wenting Zhao Jack Hessel Sean Ren Claire Cardie Yejin Choi 📝huggingface.co/papers/2409.03… 🔗wildvisualizer.com/embeddings/eng… 1/7

Introducing Meta’s Open Materials 2024 (OMat24) Dataset and Models! All under permissive open licenses for commercial and non-commercial use! Paper: arxiv.org/abs/2410.12771 Dataset: huggingface.co/datasets/fairc… Models: huggingface.co/fairchem/OMAT24 🧵1/x

How many reasoning tokens does OpenAI o1 use? It turns out they are almost always multiples of 64 (99+% of the time in 100K collected turns)🤔Could it be that the model only uses multiples of 64 tokens to think? Or maybe OpenAI rounds the token count in the returned usage? 1/4

Yuntian Deng from University of Waterloo presented at CILVR on internalizing reasoning in language models. His team's finetuning approach enabled GPT-2 Small to solve 20x20 multiplication with 99.5% accuracy, while standard training couldn't go beyond 4x4. Demo up at huggingface.co/spaces/yuntian…

New paper out with FAIR(+FAIR-Chemistry): Adjoint Sampling: Highly Scalable Diffusion Samplers via Adjoint Matching We present a scalable method for sampling from unnormalized densities beyond classical force fields. 📄: arxiv.org/abs/2504.11713

We've open sourced Adjoint Sampling! It's part of a bundled release showcasing FAIR's research and open source commitment to AI for science. github.com/facebookresear… x.com/AIatMeta/statu…

Can we build an operating system entirely powered by neural networks? Introducing NeuralOS: towards a generative OS that directly predicts screen images from user inputs. Try it live: neural-os.com Paper: huggingface.co/papers/2507.08… Inspired by Andrej Karpathy's vision. 1/5

Our new paper Digital Discovery is out💡Kyunghyun Cho Angelica Chen Do LLMs capture intrinsic chemistry, beyond string representations? Across SMILES & IUPAC: ❌ SOTA models: very low consistency (<1%) ⚖️ 1-to-1 mapped finetune ↑ consistency, not accuracy Read: pubs.rsc.org/en/Content/Art…

🚀New dataset release: WildChat-4.8M 4.8M real user-ChatGPT conversations collected from our public chatbots: - 122K from reasoning models (o1-preview, o1-mini): represent real uses in the wild and very costly to collect - 2.5M from GPT-4o 🔗 hf.co/datasets/allen… (1/4)