Skip to content

K-QA #45

Description

@warner-benjamin

Implement verifiers env for K-QA

Implement a verifiers environment for the K-QA datasets. Use the authors’ question, system, and LLM-as-judge prompts if provided. Only create a training dataset if a training split exists.

Paper: https://arxiv.org/abs/2401.14493
Dataset / Project: https://github.com/Itaymanes/K-QA

Format & Env

  • Single-turn open-ended clinical Q&A:
  • Use vf.SingleTurnEnv + FActScore +vf.JudgeRubric()

Activity

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

Labels

No labels
No labels

Type

No type

Projects

No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions