update (microsoft#2178)

Co-authored-by: AnonymousRepoSub <“[email protected]” >

website/blog/2023-12-23-AgentOptimizer/index.mdx

@@ -12,10 +12,12 @@ tags: [LLM, research]
 **TL;DR:**
 Introducing **AgentOptimizer**, a new class for training LLM agents in the era of LLMs as a service.
 **AgentOptimizer** is able to prompt LLMs to iteratively optimize function/skills of AutoGen agents according to the historical conversation and performance.
-Checkout one implementation for **AgentOptimizer** on [MATH](https://github.com/hendrycks/math) dataset
-[here](https://github.com/microsoft/autogen/blob/main/notebook/agentchat_agentoptimizer.ipynb).
 
-More information could be found in the [paper](https://arxiv.org/abs/2402.11359).
+More information could be found in:
+
+**Paper**: https://arxiv.org/abs/2402.11359.
+
+**Notebook**: https://github.com/microsoft/autogen/blob/main/notebook/agentchat_agentoptimizer.ipynb.
 
 ## Introduction
 In the traditional ML pipeline, we train a model by updating its weights according to the loss on the training set, while in the era of LLM agents, how should we train an agent?