From 3e9d65dff333a86ccad3f03414a6dd45f3f48a3a Mon Sep 17 00:00:00 2001 From: Jiarui Xu <39042389+jxudata@users.noreply.github.com> Date: Mon, 5 Oct 2026 11:31:24 -0700 Subject: [PATCH] Keep the README overview and link the website to GitHub --- README.md | 4 +++- materials/website/index.html | 4 ++-- 2 files changed, 5 insertions(+), 3 deletions(-) diff --git a/README.md b/README.md index 08f0b4f..3efc63d 100644 --- a/README.md +++ b/README.md @@ -8,7 +8,7 @@ [Python notebook](https://colab.research.google.com/github/Affirm/tabular-foundation-models-tutorial/blob/main/materials/notebooks/01_tabicl_primer.ipynb) · [Our guide](#our-interactive-guide) · [Model landscape](#model-landscape) · [Learning resources](#tutorials-and-learning-resources) · [Conference resources](#conference-and-workshop-resources) -Table as Prompt: An Interactive Guide to Tabular Foundation Models - accepted for presentation at the [NeurIPS 2026 Education Track](https://neurips.cc/Conferences/2026/CallforEducationalResources). [Read the paper (PDF, submission version)](materials/website/paper.pdf). +**Table as Prompt: An Interactive Guide to Tabular Foundation Models** - accepted for presentation at the [NeurIPS 2026 Education Track](https://neurips.cc/Conferences/2026/CallforEducationalResources). [Read the paper (PDF, submission version)](materials/website/paper.pdf). ## What are tabular foundation models? @@ -20,6 +20,8 @@ The broader goal is reusable prediction for structured data. The field now cover ## Our interactive guide +![Tabular Foundation Models Tutorial: Interactive Guide, Papers and Code, and Benchmarks. Tabular in-context learning uses labeled examples and a new row's features as inputs to a pretrained Transformer with fixed weights to predict the new row's label.](assets/tabular-tutorial-overview.png) + Open the [guide](https://affirm.github.io/tabular-foundation-models-tutorial/) in a modern browser. Follow the sequence from tabular prediction and adaptation through PFN theory, TabICLv2 training, and inference. Then try the browser playground and inspect the model's intermediate computations. Basic supervised learning is enough to get started. TabICLv2 is the worked example. The guide distinguishes explanatory simulations from real model execution; browser inference runs locally. For a Python exercise, use the [notebook](materials/notebooks/01_tabicl_primer.ipynb) or [open it in Colab](https://colab.research.google.com/github/Affirm/tabular-foundation-models-tutorial/blob/main/materials/notebooks/01_tabicl_primer.ipynb). [Setup instructions](materials/README.md) cover the notebook environment. diff --git a/materials/website/index.html b/materials/website/index.html index 75a5304..8900b6c 100644 --- a/materials/website/index.html +++ b/materials/website/index.html @@ -34,11 +34,11 @@ NeurIPS 2026 · Education Track - + - Website + GitHub Tabular Foundation Models