Skip to content

Conversation

@trungthanhnguyen0502
Copy link
Collaborator

@trungthanhnguyen0502 trungthanhnguyen0502 commented Feb 17, 2025

Update scoring mechanism:

  • extract number with regex. Compute float difference if both groundtruth and miner_answer contain only one number value
  • extract final answer by LLM with EXTRACT_ANSWER_PROMPT.
  • Remove task catergory: logic and gen-code

@trungthanhnguyen0502 trungthanhnguyen0502 changed the base branch from main to pre-release February 17, 2025 08:22
@LVH-Tony LVH-Tony merged commit 61dbc57 into pre-release Feb 17, 2025
trungthanhnguyen0502 pushed a commit that referenced this pull request Apr 22, 2025
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants