Update app.py

app.py

@@ -97,17 +97,29 @@ with demo:
     with gr.Tabs(elem_classes="tab-buttons") as tabs:
         with gr.TabItem("🏅 ContextualBenchmark", elem_id="llm-benchmark-tab-table", id=0):
             # leaderboard = init_leaderboard(LEADERBOARD_DF)
-            df_ic = pd.read_table("image_identification.tsv")
+            df_ii = pd.read_table("image_identification.tsv")
             with gr.Blocks():
                 with gr.Row():
-                    gr.Markdown("""# Image identification result
+                    gr.Markdown("""# Image identification (Binary Accuracy)
                     """)
                 with gr.Column():
                     leaderboard_df = gr.components.DataFrame(
-                        value=df_ic,
+                        value=df_ii,
                         datatype=["markdown", "number", "number", "number", "number", "number", "number", "number"]
                     )
 
+            df_ie = pd.read_table("image_explanation.tsv")
+            with gr.Blocks():
+                with gr.Row():
+                    gr.Markdown("""# Image explanation (GPT-4 Rating)
+                    """)
+                with gr.Column():
+                    leaderboard_df = gr.components.DataFrame(
+                        value=df_ie,
+                        datatype=["markdown", "number", "number", "number", "number", "number", "number", "number"]
+                    )
+
+
 
         with gr.TabItem("📝 About", elem_id="llm-benchmark-tab-table", id=2):
             gr.Markdown(LLM_BENCHMARKS_TEXT, elem_classes="markdown-text")