updated index

Author: Gianni Crivello

nbs/index.ipynb

@@ -30,6 +30,7 @@
     "In the rapidly evolving landscape of Large Language Models (LLMs), we're constantly discovering new capabilities and limitations. And more common than not, we are also discovering the limits of our new capabilities! One area that's been gaining traction in industrial applications is structured generation - the ability to produce outputs in standardized formats like JSON or XML. But what if these format restrictions come at a cost? A fascinating new study titled \"Let Me Speak Freely? A Study on the Impact of Format Restrictions on Performance of Large Language Models\" dives deep into this question, uncovering some surprising findings that could reshape how we approach LLM implementations in real-world scenarios.\n",
     "\n",
     "## The Dilemma: Structure vs. Performance\n",
+    "\n",
     "At the heart of this study is one question: Do format-restricting instructions affect the quality of LLMs' generated content? The researchers set out to investigate whether the constraints we impose for the sake of parsability and consistency might actually be hampering the reasoning abilities of these powerful models.\n",
     "\n",
     "## Key Findings\n",