From 08cfacdf22a549ef7c6087a555b11857cea2cbcb Mon Sep 17 00:00:00 2001
From: Aman Madaan <amadaan@cs.cmu.edu>
Date: Wed, 29 Nov 2023 14:33:14 -0500
Subject: [PATCH] [feat] add colab for visual self-refine

---
 README.md | 32 ++++++++++++++++++++++----------
 1 file changed, 22 insertions(+), 10 deletions(-)

diff --git a/README.md b/README.md
index c8ff7f7..d836a79 100644
--- a/README.md
+++ b/README.md
@@ -15,21 +15,33 @@ _With Self-Refine, LLMs can generate feedback on their work, use it to improve t
 <!-- DON'T EDIT THIS SECTION, INSTEAD RE-RUN doctoc TO UPDATE -->
 **Table of Contents**
 
-- [Setup](#setup)
-- [Getting Started with Acronym Generation](#getting-started-with-acronym-generation)
-- [Dialogue Response Generation](#dialogue-response-generation)
-- [Code Readability Improvement](#code-readability-improvement)
-- [Commongen](#commongen)
-- [GSM-8k](#gsm-8k)
-- [Yelp](#yelp)
-- [PIE](#pie)
-- [General setup](#general-setup)
-- [Citation](#citation)
+- [Self-Refine: Iterative Refinement with Self-Feedback](#self-refine-iterative-refinement-with-self-feedback)
+  - [Updates](#updates)
+  - [Setup](#setup)
+  - [Getting Started with Acronym Generation](#getting-started-with-acronym-generation)
+  - [Dialogue Response Generation](#dialogue-response-generation)
+  - [Code Readability Improvement](#code-readability-improvement)
+  - [Commongen](#commongen)
+  - [GSM-8k](#gsm-8k)
+  - [Yelp](#yelp)
+  - [PIE](#pie)
+  - [General setup](#general-setup)
+  - [Citation](#citation)
 
 <!-- END doctoc generated TOC please keep comment here to allow auto update -->
 
 <hr>
 
+## Updates
+
+- **Nov 2023**: Added [visual self-refine examples](docs/visual_self_refine_examples/) and [colabs](colabs/Visual-Self-Refine-GPT4V.ipynb). Use GPT4-V to write tikz code for diagrams, and improve them iteratively.
+
+<p>
+  <img src="docs/visual_self_refine_examples/stokes__theorem.gif" alt="Visual Self-Refine Example 1" width="300" height="300" />
+  <img src="docs/visual_self_refine_examples/unicorn.gif" alt="Visual Self-Refine Example 2" width="300" height="300" />
+</p>
+
+
 ## Setup
 
 * We use [prompt-lib](https://github.com/reasoning-machines/prompt-lib) for querying LLMs. You can install `prompt-lib` by running the following commands: