Update Arena Hard tasks docstrings

src/distilabel/steps/tasks/benchmarks/arena_hard.py

@@ -24,7 +24,9 @@
 
 
 class ArenaHard(Task):
-    """This `Task` is based on the "From Live Data to High-Quality Benchmarks: The
+    """Evaluates two assistant responses using an LLM as judge.
+
+    This `Task` is based on the "From Live Data to High-Quality Benchmarks: The
     Arena-Hard Pipeline" paper that presents Arena Hard, which is a benchmark for
     instruction-tuned LLMs that contains 500 challenging user queries. GPT-4 is used
     as the judge to compare the model responses against a baseline model, which defaults
@@ -145,7 +147,9 @@ def format_output(
 
 
 class ArenaHardResults(GlobalStep):
-    """This `Step` is based on the "From Live Data to High-Quality Benchmarks: The
+    """Process Arena Hard results to calculate the ELO scores.
+
+    This `Step` is based on the "From Live Data to High-Quality Benchmarks: The
     Arena-Hard Pipeline" paper that presents Arena Hard, which is a benchmark for
     instruction-tuned LLMs that contains 500 challenging user queries. This step is
     a `GlobalStep` that should run right after the `ArenaHard` task to calculate the
@@ -155,6 +159,10 @@ class ArenaHardResults(GlobalStep):
         Arena-Hard-Auto has the highest correlation and separability to Chatbot Arena
         among popular open-ended LLM benchmarks.
 
+    Input columns:
+        - evaluation (`str`): The evaluation of the responses generated by the LLMs.
+        - score (`str`): The score extracted from the evaluation.
+
     References:
         - [From Live Data to High-Quality Benchmarks: The Arena-Hard Pipeline](https://lmsys.org/blog/2024-04-19-arena-hard/)
         - [`arena-hard-auto`](https://github.com/lm-sys/arena-hard-auto/tree/main)