diff --git a/README.md b/README.md
index 8e0dfb8..6edcbb0 100644
--- a/README.md
+++ b/README.md
@@ -6,6 +6,8 @@
   <img src="https://raw.githubusercontent.com/declare-lab/instruct-eval/main/docs/logo.png" alt="" width="300" height="300">
 </p>
 
+> 🔥 If you are interested in IQ testing LLMs, check out our new work: [PuzzleVQA](https://github.com/declare-lab/puzzle-reasoning)
+
 > 📣 Introducing Resta: **Safety Re-alignment of Language Models**. [**Paper**](https://arxiv.org/abs/2402.11746) [**Github**](https://github.com/declare-lab/resta)
 
 > 📣 **Red-Eval**, the benchmark for **Safety** Evaluation of LLMs has been added: [Red-Eval](https://github.com/declare-lab/instruct-eval/tree/main/red-eval)