-
Notifications
You must be signed in to change notification settings - Fork 2
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Corr2Cause #3
Corr2Cause #3
Conversation
Answer if hypothesis is correct given the premise Again use the generation adapter and the text input is premise, hypothesis and then then 2 references is the yes or no |
src/helm/benchmark/run_specs.py
Outdated
You will be given a scenario which includes a premise and a hypothesis. It is your task to determine whether the hypothesis is correct given the premise. | ||
You will be presented with two choices yes or no for each scenario. Please select the correct choice. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Maybe: "Give a scenario with a premise and a hypothesis, determine if the hypothesis can be inferred from the premise."
The verb "infer" seems most faithful to the original paper... alternatives are "suggest" and "entail".
csv_reader = csv.reader(f) | ||
next(csv_reader) | ||
for label, question in csv_reader: | ||
# formatted_input = f"{question}\nFirst Option: 0\nSecond Option: 1" |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Make sure the formatting is correct here.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
There was an issue where answer was always a again so just fixed
The only thing we might want to wordsmith more is the prompt in
get_corr2cause_spec()