update display

src/display/about.py

@@ -11,24 +11,48 @@ class Task:
 # Init: to update with your specific keys
 class Tasks(Enum):
     # task_key in the json file, metric_key in the json file, name to display in the leaderboard 
-    task0 = Task("task_name1", "metric_name", "First task")
-    task1 = Task("task_name2", "metric_name", "Second task")
+    task0 = Task("finance_bench", "accuracy", "FinanceBench")
+    task1 = Task("legal_confidentiality", "accuracy", "Legal Confidentiality")
+    task2 = Task("writing-prompts", "coherence", "Writing Prompts")
+    task3 = Task("customer-support", "engagement", "Customer Support Dialogue")
+    task4 = Task("toxic-prompts", "toxicity", "Toxic Prompts")
+    task5 = Task("enterprise-pii", "accuracy", "Enterprise PII")
 
 
 # Your leaderboard name
-TITLE = """<h1 align="center" id="space-title">Demo leaderboard</h1>"""
+TITLE = """<h1 align="center" id="space-title">Patronus AI leaderboard</h1>"""
 
 # What does your leaderboard evaluate?
 INTRODUCTION_TEXT = """
-Intro text
+This leaderboard evaluates the performance of models on real-world enterprise use cases. 
 """
 
 # Which evaluations are you running? how can people reproduce what you have?
 LLM_BENCHMARKS_TEXT = f"""
 ## How it works
 
+## Tasks
+1. FinanceBench: The task measures the ability to answer financial questions given the context. 
+
+2. Legal Confidentiality: The task measures the ability of LLMs to reason over legal causes. The model is prompted 
+to return yes/no as an answer to the question. 
+
+3. Writing Prompts: This task evaluates the story-writing and creative abilities of the LLM. 
+
+4. Customer Support Dialogue: This task evaluates the ability of the LLM to answer a customer support question 
+given some product information and conversational history. 
+
+5. Toxic Prompts: This task evaluates the safety of the model by using prompts that can elicit harmful information 
+from LLMs.
+
+6. Enterprise PII: This task evaluates the business safety of the model by using prompts to elicit business-sensitive information from LLMs.
+
 ## Reproducibility
-To reproduce our results, here is the commands you can run:
+All of our datasets are closed-source. We provide a validation set with 5 examples for each of the tasks.
+
+To reproduce the results on the validation set, run:
+
+
 
 """