leaderboard update

app.py

@@ -12,7 +12,7 @@ from src.about import (
 
 data = {
     "Method": [
-        "Handwritten TAG",
+        "Handwritten LOTUS (llama-3.1-70B)",
         "Zero-shot Text2SQL (llama-3.1-70B)",
         "Zero-shot Text2SQL + LM Generation (llama-3.1-70B)",
         "RAG (E5)",
@@ -27,7 +27,17 @@ leaderboard_df = pd.DataFrame(data)
 leaderboard_df = leaderboard_df.sort_values(
     "Execution Accuracy", ascending=False
 ).reset_index(drop=True)
-leaderboard_df.insert(0, "Rank", leaderboard_df.index + 1)
+leaderboard_df.insert(0, "Rank", leaderboard_df.index + 2)
+leaderboard_df.loc[0, "Rank"] = None
+
+def highlight_row(row):
+    if pd.isna(row["Rank"]):  # First row
+        return ["background-color: #d4edda; font-weight: bold;" for _ in row]
+    return [""] * len(row)
+
+
+# Apply the style
+leaderboard_df = leaderboard_df.style.apply(highlight_row, axis=1)
 
 
 def hyperlink_method(method):
@@ -87,7 +97,7 @@ with gr.Blocks() as demo:
                 )
 
             # Section 2: Submission Frequency
-            with gr.Accordion("2️⃣ Submission Frequency", open=False):
+            with gr.Accordion("2️⃣ Submission Frequency", open=True):
                 gr.Markdown(
                     """
                     - Submissions are accepted **once a month** to ensure sufficient evaluation bandwidth.
@@ -96,17 +106,17 @@ with gr.Blocks() as demo:
                 )
 
             # Section 3: How to Upload Materials
-            with gr.Accordion("3️⃣ How to Upload Materials", open=False):
+            with gr.Accordion("3️⃣ How to Upload Materials", open=True):
                 gr.Markdown(
                     """
                     Follow these steps to upload your materials:
                     1. Compress all files in the code into a single `.zip` file, or provide a public repository to refer to.
-                    2. Email the `.zip` file or repositoty link to our email [email].
+                    2. Email the `.zip` file or repositoty link to our email tagbenchmark@gmail.com.
                     """
                 )
 
             # Section 4: Submission Process
-            with gr.Accordion("4️⃣ Submission Process", open=False):
+            with gr.Accordion("4️⃣ Submission Process", open=True):
                 gr.Markdown(
                     """
                     After uploading your materials:
@@ -122,7 +132,7 @@ with gr.Blocks() as demo:
             gr.Markdown(
                 """
                 <div style="text-align: center; margin-top: 2rem;">
-                    For further assistance, reach out to [email] with questions.
+                    For further assistance, reach out to tagbenchmark@gmail.com with questions.
                 </div>
                 """
             )

src/about.py

@@ -31,7 +31,7 @@ Intro text
 # Which evaluations are you running? how can people reproduce what you have?
 LLM_BENCHMARKS_TEXT = f"""
 ## What does the TAG leaderboard evaluate?
-In this leaderboard, you'll find execution accuracy comparisons of table question answering approaches on [TAG-Bench] (https://github.com/TAG-Research/TAG-Bench/tree/main). TAG-Bench contains complex queries requiring world knowledge or semantic reasoning that goes beyond the information explicitly available in the database.
+In this leaderboard, you'll find execution accuracy comparisons of table question answering approaches on [TAG-Bench](https://github.com/TAG-Research/TAG-Bench/tree/main). TAG-Bench contains complex queries requiring world knowledge or semantic reasoning that goes beyond the information explicitly available in the database.
 
 ## How is accuracy measured?
 Execution accuracy is measured as the number of exact matches to our annotated ground truth answers which are hand-labeled by experts.