Skip to content

fix(rm): deepscaler silently zeroes all rewards for non-thinking models #2219

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.

Sign up for GitHub

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

Open
acmore wants to merge 1 commit into
base: main
Choose a base branch
from
Open

fix(rm): deepscaler silently zeroes all rewards for non-thinking models #2219

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
5 changes: 4 additions & 1 deletion slime/rollout/rm_hub/deepscaler.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -7,7 +7,10 @@ def get_deepscaler_rule_based_reward(response, label):
elif "###Response" in response:
model_solution = response.split("###Response")[1]
else:
return 0
# Non-thinking models emit neither marker; grade the whole response
# instead of silently returning 0 (which zeroes every reward and
# kills the GRPO signal without any visible error).
model_solution = response

model_answer = extract_answer(model_solution)
if model_answer is None:
Expand Down
Loading