Update space

src/about.py

@@ -57,11 +57,10 @@ TITLE = """<h1 align="center" id="space-title">Decentralized Arena</h1>"""
 
 # What does your leaderboard evaluate?
 INTRODUCTION_TEXT = """
-# Introduction
-TL;DR: We release Decentralized Arena that automates and scales “Chatbot Arena” for LLM evaluation across various 
-fine-grained dimensions (e.g., math – algebra, geometry, probability; logical reasoning, social reasoning, 
-biology, chemistry, …). The evaluation is decentralized and democratic, with all LLMs participating 
-in evaluating others. It achieves a 97\% correlation with Chatbot Arena's overall rankings, while being fully transparent and reproducible.
+Decentralized Arena automates and scales "Chatbot Arena" for LLM evaluation across various fine-grained dimensions 
+(e.g., math – algebra, geometry, probability; logical reasoning, social reasoning, biology, chemistry, …). 
+The evaluation is decentralized and democratic, with all LLMs participating in evaluating others. 
+It achieves a 95\% correlation with Chatbot Arena's overall rankings, while being fully transparent and reproducible.
 """
 
 # Which evaluations are you running? how can people reproduce what you have?

src/display/utils.py

@@ -64,26 +64,26 @@ auto_eval_column_dict.append(["score_sd", ColumnContent, field(default_factory=l
 auto_eval_column_dict.append(["rank", ColumnContent, field(default_factory=lambda: ColumnContent("Rank", "number", True))])
 
 # fine-graine dimensions
-auto_eval_column_dict.append(["score_overall", ColumnContent, field(default_factory=lambda: ColumnContent("Overall", "number", True))])
-auto_eval_column_dict.append(["score_math_algebra", ColumnContent, field(default_factory=lambda: ColumnContent("Math (Algebra)", "number", True))])
-auto_eval_column_dict.append(["score_math_geometry", ColumnContent, field(default_factory=lambda: ColumnContent("Math (Geometry)", "number", True))])
-auto_eval_column_dict.append(["score_math_probability", ColumnContent, field(default_factory=lambda: ColumnContent("Math (Probability)", "number", True))])
-auto_eval_column_dict.append(["score_reason_logical", ColumnContent, field(default_factory=lambda: ColumnContent("Logical Reasoning", "number", True))])
-auto_eval_column_dict.append(["score_reason_social", ColumnContent, field(default_factory=lambda: ColumnContent("Social Reasoning", "number", True))])
-
-auto_eval_column_dict.append(["sd_overall", ColumnContent, field(default_factory=lambda: ColumnContent("SD Overall", "number", True))])
-auto_eval_column_dict.append(["sd_math_algebra", ColumnContent, field(default_factory=lambda: ColumnContent("SD Math (Algebra)", "number", True))])
-auto_eval_column_dict.append(["sd_math_geometry", ColumnContent, field(default_factory=lambda: ColumnContent("SD Math (Geometry)", "number", True))])
-auto_eval_column_dict.append(["sd_math_probability", ColumnContent, field(default_factory=lambda: ColumnContent("SD Math (Probability)", "number", True))])
-auto_eval_column_dict.append(["sd_reason_logical", ColumnContent, field(default_factory=lambda: ColumnContent("SD Logical Reasoning", "number", True))])
-auto_eval_column_dict.append(["sd_reason_social", ColumnContent, field(default_factory=lambda: ColumnContent("SD Social Reasoning", "number", True))])
-
-auto_eval_column_dict.append(["rank_overall", ColumnContent, field(default_factory=lambda: ColumnContent("Rank Overall", "number", True))])
-auto_eval_column_dict.append(["rank_math_algebra", ColumnContent, field(default_factory=lambda: ColumnContent("Rank Math (Algebra)", "number", True))])
-auto_eval_column_dict.append(["rank_math_geometry", ColumnContent, field(default_factory=lambda: ColumnContent("Rank Math (Geometry)", "number", True))])
-auto_eval_column_dict.append(["rank_math_probability", ColumnContent, field(default_factory=lambda: ColumnContent("Rank Math (Probability)", "number", True))])
-auto_eval_column_dict.append(["rank_reason_logical", ColumnContent, field(default_factory=lambda: ColumnContent("Rank Logical Reasoning", "number", True))])
-auto_eval_column_dict.append(["rank_reason_social", ColumnContent, field(default_factory=lambda: ColumnContent("Rank Social Reasoning", "number", True))])
+auto_eval_column_dict.append(["score_overall", ColumnContent, field(default_factory=lambda: ColumnContent("Score (Overall)", "number", True))])
+auto_eval_column_dict.append(["score_math_algebra", ColumnContent, field(default_factory=lambda: ColumnContent("Score (Math Algebra)", "number", True))])
+auto_eval_column_dict.append(["score_math_geometry", ColumnContent, field(default_factory=lambda: ColumnContent("Score (Math Geometry)", "number", True))])
+auto_eval_column_dict.append(["score_math_probability", ColumnContent, field(default_factory=lambda: ColumnContent("Score (Math Probability)", "number", True))])
+auto_eval_column_dict.append(["score_reason_logical", ColumnContent, field(default_factory=lambda: ColumnContent("Score (Logical Reasoning)", "number", True))])
+auto_eval_column_dict.append(["score_reason_social", ColumnContent, field(default_factory=lambda: ColumnContent("Score (Social Reasoning)", "number", True))])
+
+auto_eval_column_dict.append(["sd_overall", ColumnContent, field(default_factory=lambda: ColumnContent("Std dev(Overall)", "number", True))])
+auto_eval_column_dict.append(["sd_math_algebra", ColumnContent, field(default_factory=lambda: ColumnContent("Std dev (Math Algebra)", "number", True))])
+auto_eval_column_dict.append(["sd_math_geometry", ColumnContent, field(default_factory=lambda: ColumnContent("Std dev (Math Geometry)", "number", True))])
+auto_eval_column_dict.append(["sd_math_probability", ColumnContent, field(default_factory=lambda: ColumnContent("Std dev (Math Probability)", "number", True))])
+auto_eval_column_dict.append(["sd_reason_logical", ColumnContent, field(default_factory=lambda: ColumnContent("Std dev (Logical Reasoning)", "number", True))])
+auto_eval_column_dict.append(["sd_reason_social", ColumnContent, field(default_factory=lambda: ColumnContent("Std dev (Social Reasoning)", "number", True))])
+
+auto_eval_column_dict.append(["rank_overall", ColumnContent, field(default_factory=lambda: ColumnContent("Rank (Overall)", "number", True))])
+auto_eval_column_dict.append(["rank_math_algebra", ColumnContent, field(default_factory=lambda: ColumnContent("Rank (Math Algebra)", "number", True))])
+auto_eval_column_dict.append(["rank_math_geometry", ColumnContent, field(default_factory=lambda: ColumnContent("Rank (Math Geometry)", "number", True))])
+auto_eval_column_dict.append(["rank_math_probability", ColumnContent, field(default_factory=lambda: ColumnContent("Rank (Math Probability)", "number", True))])
+auto_eval_column_dict.append(["rank_reason_logical", ColumnContent, field(default_factory=lambda: ColumnContent("Rank (Logical Reasoning)", "number", True))])
+auto_eval_column_dict.append(["rank_reason_social", ColumnContent, field(default_factory=lambda: ColumnContent("Rank (Social Reasoning)", "number", True))])
                              
 
 for task in Tasks: