diff --git a/README.md b/README.md
index 6e41332..4f1cbb2 100644
--- a/README.md
+++ b/README.md
@@ -55,7 +55,7 @@ Once published on arXiv:
 
 1. **Quick Overview**: Read `executive_summary.md` (2 pages)
 2. **Technical Summary**: Read `two_page_summary.tex` (2 pages, compile to PDF)
-3. **Full Paper**: Read `main.tex` (24 pages, compile to PDF)
+3. **Full Paper**: Read `main.tex` (25 pages, compile to PDF)
 4. **Try It Yourself**: Visit the [experiments repository](https://github.com/sqrtspace/sqrtspace-experiments)
 
 ## Contact
diff --git a/figures/ollama_sqrt_validation.png b/figures/ollama_sqrt_validation.png
new file mode 100644
index 0000000..3ae4ea5
Binary files /dev/null and b/figures/ollama_sqrt_validation.png differ
diff --git a/main.tex b/main.tex
index feddd2f..f46efcf 100644
--- a/main.tex
+++ b/main.tex
@@ -505,6 +505,13 @@ Chunked $\sqrt{n}$ & 54.10 $\pm$ 2.71s & 2.41 MB & 122 & 18.3× \\
 
 The 18.3× slowdown aligns more closely with theoretical predictions than our simulated results, demonstrating that real models exhibit the expected space-time tradeoffs when processing is dominated by model inference rather than memory bandwidth.
 
+\begin{figure}[htbp]
+\centering
+\includegraphics[width=0.75\textwidth]{figures/ollama_sqrt_validation.png}
+\caption{Validation that our Ollama context chunking follows the theoretical $\sqrt{n}$ pattern. For 14,750 characters of input, we use 122 chunks of 121 characters each, precisely following $\sqrt{n}$ chunking.}
+\label{fig:ollama_sqrt}
+\end{figure}
+
 \begin{figure}[htbp]
 \centering
 \includegraphics[width=0.95\textwidth]{figures/ollama_spacetime_results.png}
diff --git a/ubiquity.pdf b/ubiquity.pdf
index bd6dba6..f9d1e0b 100644
Binary files a/ubiquity.pdf and b/ubiquity.pdf differ