diff --git a/backend/tests/evaluation/compare_scores.py b/backend/tests/evaluation/compare_scores.py
index 0c06e53..efc8b89 100644
--- a/backend/tests/evaluation/compare_scores.py
+++ b/backend/tests/evaluation/compare_scores.py
@@ -1,4 +1,5 @@
 import sys
+from pathlib import Path
 
 import pandas as pd
 
@@ -25,6 +26,21 @@ def read_markdown_table(file_path):
     return df
 
 
+def get_emoji(difference):
+    if abs(difference) < 1:
+        return "✅"  # Green checkmark for small differences
+    elif difference > 0:
+        if difference > 5:
+            return "🚀"  # Rocket for significant improvements
+        else:
+            return "📈"  # Chart with upwards trend for improvements
+    else:
+        if difference < -5:
+            return "⚠️"  # Warning sign for significant regressions
+        else:
+            return "📉"  # Chart with downwards trend for regressions
+
+
 def compare_scores(current_scores_path, main_scores_path):
     # Read current scores
     try:
@@ -59,15 +75,23 @@ def compare_scores(current_scores_path, main_scores_path):
         (merged_df["Score_current"] - merged_df["Score_main"]) / merged_df["Score_main"] * 100
     ).round(2)
 
+    # Add emoji column
+    merged_df["Emoji"] = merged_df["Difference"].apply(get_emoji)
+
     # Prepare output dataframe
-    output_df = merged_df[["Tag", "Score_current", "Score_main", "Difference"]]
-    output_df.columns = ["Tag", "Current Score", "Main Score", "Difference (%)"]
+    output_df = merged_df[["Tag", "Score_current", "Score_main", "Difference", "Emoji"]]
+    output_df.columns = ["Tag", "PR Score", "Main Branch Score", "Difference (%)", "Status"]
     output_df = output_df.fillna("N/A")
 
     # Write to markdown file
     with open("comparison.md", "w") as f:
         f.write("## LLM Evaluation Score Comparison\n\n")
-        output_df.to_markdown(index=False, buf=f, mode="w")
+        f.write("| Tag | PR Score | Main Branch Score | Difference (%) | Status |\n")
+        f.write("|-----|----------|-------------------|----------------|--------|\n")
+        for _, row in output_df.iterrows():
+            f.write(
+                f"| {row['Tag']} | {row['PR Score']:.2f} | {row['Main Branch Score']:.2f} | {row['Difference (%)']:+.2f}% | {row['Status']} |\n"
+            )
 
 
 if __name__ == "__main__":