SciCode / README.md
akshathmangudi's picture
Re-write Parquet with smaller row groups for HF dataset viewer
526f6cb verified
metadata
dataset_info:
  features:
    - name: benchmark
      dtype: string
    - name: artifact_type
      dtype: string
    - name: problem_id
      dtype: string
    - name: test_id
      dtype: string
    - name: variables
      dtype: string
  splits:
    - name: train
      num_bytes: 1889380696
      num_examples: 1082
  download_size: 820277839
  dataset_size: 1889380696
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*
license: mit
task_categories:
  - question-answering
language:
  - en
tags:
  - code
pretty_name: SciCode
size_categories:
  - 1K<n<10K

Dataset Card for Dataset Name

Official Description (from the authors): Since language models (LMs) now outperform average humans on many challenging tasks, it has become increasingly difficult to develop challenging, high-quality, and realistic evaluations. We address this issue by examining LMs' capabilities to generate code for solving real scientific research problems. Incorporating input from scientists and AI researchers in 16 diverse natural science sub-fields, including mathematics, physics, chemistry, biology, and materials science, we created a scientist-curated coding benchmark, SciCode. The problems in SciCode naturally factorize into multiple subproblems, each involving knowledge recall, reasoning, and code synthesis. In total, SciCode contains 338 subproblems decomposed from 80 challenging main problems. It offers optional descriptions specifying useful scientific background information and scientist-annotated gold-standard solutions and test cases for evaluation. Claude3.5-Sonnet, the best-performing model among those tested, can solve only 4.6% of the problems in the most realistic setting. We believe that SciCode demonstrates both contemporary LMs' progress towards becoming helpful scientific assistants and sheds light on the development and evaluation of scientific AI in the future.

This repository contains the ground truth artifacts that's needed for LightEval benchmarks.

The original SciCode numerical evaluation artifacts are provided in raw/raw_ground.h5 for reproducibility and parity with the original SciCode evaluation pipeline.

This dataset uses a single split (train) as it represents a complete set of SciCode numerical evaluation artifacts rather than training data.

Dataset Details

Dataset Sources [optional]

Dataset Card Authors

The original authors of SciCode benchmark and Akshath Mangudi for providing the ground truth artifact.