Revert homepage simplification - keep original design

category_page_builder.py

@@ -12,15 +12,8 @@ CATEGORY_DIAGRAM_MAP = {
 
 def build_category_page(CATEGORY_NAME, PAGE_DESCRIPTION):
     with gr.Column(elem_id="page-content-wrapper"):
-        try:
-            validation_df, validation_tag_map = get_full_leaderboard_data("validation")
-            test_df, test_tag_map = get_full_leaderboard_data("test")
-        except Exception as e:
-            print(f"Error loading data for {CATEGORY_NAME}: {e}")
-            validation_df = pd.DataFrame()
-            validation_tag_map = {}
-            test_df = pd.DataFrame()
-            test_tag_map = {}
+        validation_df, validation_tag_map = get_full_leaderboard_data("validation")
+        test_df, test_tag_map = get_full_leaderboard_data("test")
         with gr.Row(elem_id="intro-row"):
 
             with gr.Column(scale=1):

main_page.py

@@ -33,33 +33,21 @@ def build_page():
 
     # --- Leaderboard Display Section ---
     gr.Markdown("---")
-    gr.HTML('<h2>OpenHands Index Leaderboards</h2>', elem_id="main-header")
-    
-    gr.Markdown("""
-    Select a benchmark from the navigation above to view detailed results:
-    
-    - **SWE-bench**: Software engineering benchmarks
-    - **Multi-SWE-bench**: Multi-repository software engineering tasks
-    - **SWE-bench Multimodal**: Multimodal software engineering challenges
-    - **SWT-bench**: Web testing benchmarks
-    - **Commit0**: Zero-shot code commit generation
-    - **GAIA**: General AI Assistant benchmarks
-    """)
-    
-    # Simplified - no expensive data loading on home page
-    if False:  # Disabled expensive loading
-        with gr.Tabs() as tabs:
-            with gr.Tab("Results: Test Set") as test_tab:
-                test_df, test_tag_map = get_full_leaderboard_data("test")
-                if not test_df.empty:
-                    gr.Markdown("**Test Set** results are reserved for final assessment. This helps ensure that the agent generalizes well to unseen problems.")
-                    create_leaderboard_display(
-                        full_df=test_df,
-                        tag_map=test_tag_map,
-                        category_name="Overall",
-                        split_name="test"
-                    )
-                else:
+    CATEGORY_NAME = "Overall"
+    gr.HTML(f'<h2>AstaBench {CATEGORY_NAME} Leaderboard <span style="font-weight: normal; color: inherit;">(Aggregate)</span></h2>', elem_id="main-header")
+
+    with gr.Tabs() as tabs:
+        with gr.Tab("Results: Test Set") as test_tab:
+            test_df, test_tag_map = get_full_leaderboard_data("test")
+            if not test_df.empty:
+                gr.Markdown("**Test Set** results are reserved for final assessment. This helps ensure that the agent generalizes well to unseen problems.")
+                create_leaderboard_display(
+                    full_df=test_df,
+                    tag_map=test_tag_map,
+                    category_name=CATEGORY_NAME, # Use our constant
+                    split_name="test"
+                )
+            else:
                 gr.Markdown("No data available for test split.")
         with gr.Tab("Results: Validation Set") as validation_tab:
             # 1. Load all necessary data for the "validation" split ONCE.