Update app.py

app.py

@@ -343,23 +343,29 @@ with block:
         # 2️⃣  Visualisation tab
         # ------------------------------------------------------------------
         with gr.TabItem("📊 Visualization", elem_id="probe-benchmark-tab-visualization", id=2):
+        
             gr.Markdown(
                 """## **Interactive Visualizations**  
-                Choose a benchmark type; context‑specific options will appear.""",
+                Choose a benchmark type; context-specific options will appear.""",
                 elem_classes="markdown-text",
             )
+        
+            # ── benchmark-type selector ──────────────────────────────────
             vis_benchmark_type_selector = gr.Dropdown(
                 choices=list(benchmark_specific_metrics.keys()),
                 label="🧪 Benchmark Type",
                 value=None,
             )
+        
+            # ── metric / dataset selectors (appear contextually) ─────────
             with gr.Row():
-                vis_x_metric_selector = gr.Dropdown(choices=[], label="X‑axis Metric", visible=False)
-                vis_y_metric_selector = gr.Dropdown(choices=[], label="Y‑axis Metric", visible=False)
-                vis_aspect_type_selector = gr.Dropdown(choices=[], label="Aspect", visible=False)
-                vis_dataset_selector = gr.Dropdown(choices=[], label="Dataset", visible=False)
-                vis_single_metric_selector = gr.Dropdown(choices=[], label="Metric", visible=False)
-
+                vis_x_metric_selector      = gr.Dropdown(choices=[], label="X-axis Metric", visible=False)
+                vis_y_metric_selector      = gr.Dropdown(choices=[], label="Y-axis Metric", visible=False)
+                vis_aspect_type_selector   = gr.Dropdown(choices=[], label="Aspect",       visible=False)
+                vis_dataset_selector       = gr.Dropdown(choices=[], label="Dataset",      visible=False)
+                vis_single_metric_selector = gr.Dropdown(choices=[], label="Metric",       visible=False)
+        
+            # ── method selectors (two accordions) ───────────────────────
             base_method_names = [m for m in method_names if m in base_methods]
             user_method_names = [m for m in method_names if m not in base_methods]
         
@@ -379,14 +385,15 @@ with block:
                     interactive=True,
                 )
         
+            # ── plot button & output ────────────────────────────────────
             plot_button = gr.Button("Plot")
+        
             with gr.Row(show_progress=True, variant='panel'):
                 plot_output = gr.Image(label="Plot")
-            
+        
             gr.Markdown("#### If a method name ends with **^**, it suggests potential suspicions of data leakage related to ***similarity***, ***function***, or ***family*** benchmarks.")
-
-            plot_explanation = gr.Markdown(visible=False)
-            # callbacks
+        
+            # ── callbacks ───────────────────────────────────────────────
             vis_benchmark_type_selector.change(
                 update_metric_choices,
                 inputs=[vis_benchmark_type_selector],
@@ -398,11 +405,18 @@ with block:
                     vis_single_metric_selector,
                 ],
             )
+        
+            # combine the two method lists, then call the original helper
             plot_button.click(
-                generate_plot,
+                lambda bt, base_sel, user_sel, xm, ym, asp, ds, sm: generate_plot(
+                    bt,
+                    (base_sel or []) + (user_sel or []),   # merged method list
+                    xm, ym, asp, ds, sm,
+                ),
                 inputs=[
                     vis_benchmark_type_selector,
-                    vis_method_selector,
+                    vis_method_selector_base,
+                    vis_method_selector_user,
                     vis_x_metric_selector,
                     vis_y_metric_selector,
                     vis_aspect_type_selector,