Polish: remove 5 stale 'TODO: re-render figure' comments

All 5 figures are rendered and current; the TODO comments were leftover
notes from when figures were being generated. No functional change — just
cleanup of invisible LaTeX source comments.

2 files changed, 0 insertions, 5 deletions

diff --git a/paper/main.pdf b/paper/main.pdf
index 9973b4d..bfe4793 100644
--- a/paper/main.pdf
+++ b/paper/main.pdf
diff --git a/paper/main.tex b/paper/main.tex
index a1c3bc9..d95b40c 100644
--- a/paper/main.tex
+++ b/paper/main.tex
@@ -67,7 +67,6 @@ When we compare each method to a frozen-blocks baseline matched to the same arch
 
 \begin{figure}[t]
 \centering
-% TODO: re-render figure
 \includegraphics[width=\linewidth]{figures/figure_audit_5method.png}
 \caption{Five-method audit on the 4-block $d{=}256$ pre-LayerNorm ResMLP: the field-standard pair looks superficially consistent across methods, but the diagnostic view separates trustworthy controls from walked-back methods.}
 \label{fig:audit_hero}
@@ -123,7 +122,6 @@ The architecture comparison sharpens the scope of the critique. In the terminal-
 
 \begin{figure}[t]
 \centering
-% TODO: re-render figure
 \includegraphics[width=\linewidth]{figures/figure_cross_arch_temporal_s42.png}
 \caption{Temporal and cross-architecture validation: the protocol fires early on terminal-normalized residual architectures, never fires on BP controls, and separates the activation-growth pathology from the gradient-floor pathology.}
 \label{fig:temporal_cross_arch}
@@ -131,7 +129,6 @@ The architecture comparison sharpens the scope of the critique. In the terminal-
 
 \begin{figure}[t]
 \centering
-% TODO: re-render figure
 \includegraphics[width=0.92\linewidth]{figures/fig4_penalty_rescue.pdf}
 \caption{Penalty intervention view of the two modes: penalization rescues residual-stream scale and restores a measurable but still partial deep-layer credit signal, clarifying that numerical rescue and credit-quality rescue are related but distinct.}
 \label{fig:penalty_rescue}
@@ -139,7 +136,6 @@ The architecture comparison sharpens the scope of the critique. In the terminal-
 
 \begin{figure}[t]
 \centering
-% TODO: re-render figure
 \includegraphics[width=0.92\linewidth]{figures/fig5_cross_arch_summary.pdf}
 \caption{Cross-architecture summary over ResMLP, ViT-Mini, StudentNet, and CNN: activation-growth failures recur across architectures, while gradient-floor failures appear in the terminal-normalized settings audited here.}
 \label{fig:cross_arch_summary}
@@ -173,7 +169,6 @@ A useful evaluation rule should reject the bad cases without collapsing everythi
 
 \begin{figure}[t]
 \centering
-% TODO: re-render figure
 \includegraphics[width=0.88\linewidth]{figures/fig2_decision_utility.pdf}
 \caption{Decision-utility ablation comparing the field-standard reporting pair against progressively richer diagnostic strategies: accuracy only and accuracy+$\Gamma$ walk back no audited failures, while the full protocol walks back the three silent failures.}
 \label{fig:decision_utility}