SciQ#

Overview#

SciQ is a crowdsourced science exam question dataset covering Physics, Chemistry, Biology, and other scientific domains. Most questions include supporting evidence paragraphs.

Task Description#

  • Task Type: Science Question Answering (Multiple-Choice)

  • Input: Science question with 4 answer choices

  • Output: Correct answer letter (A, B, C, or D)

  • Domains: Physics, Chemistry, Biology, Earth Science, etc.

Key Features#

  • Crowdsourced science exam questions

  • Multiple scientific domains covered

  • Supporting evidence paragraphs available

  • Tests scientific knowledge and reasoning

  • Suitable for science comprehension evaluation

Evaluation Notes#

  • Default configuration uses 0-shot evaluation

  • Uses simple multiple-choice prompting

  • Evaluates on test split

  • Simple accuracy metric

Properties#

Property

Value

Benchmark Name

sciq

Dataset ID

extraordinarylab/sciq

Paper

N/A

Tags

Knowledge, MCQ, ReadingComprehension

Metrics

acc

Default Shots

0-shot

Evaluation Split

test

Data Statistics#

Metric

Value

Total Samples

1,000

Prompt Length (Mean)

322.68 chars

Prompt Length (Min/Max)

244 / 505 chars

Sample Example#

Subset: default

{
  "input": [
    {
      "id": "a30097af",
      "content": "Answer the following multiple choice question. The entire content of your response should be of the following format: 'ANSWER: [LETTER]' (without quotes) where [LETTER] is one of A,B,C,D.\n\nCompounds that are capable of accepting electrons, such as o 2 or f2, are called what?\n\nA) antioxidants\nB) Oxygen\nC) residues\nD) oxidants"
    }
  ],
  "choices": [
    "antioxidants",
    "Oxygen",
    "residues",
    "oxidants"
  ],
  "target": "D",
  "id": 0,
  "group_id": 0,
  "metadata": {}
}

Prompt Template#

Prompt Template:

Answer the following multiple choice question. The entire content of your response should be of the following format: 'ANSWER: [LETTER]' (without quotes) where [LETTER] is one of {letters}.

{question}

{choices}

Usage#

Using CLI#

evalscope eval \
    --model YOUR_MODEL \
    --api-url OPENAI_API_COMPAT_URL \
    --api-key EMPTY_TOKEN \
    --datasets sciq \
    --limit 10  # Remove this line for formal evaluation

Using Python#

from evalscope import run_task
from evalscope.config import TaskConfig

task_cfg = TaskConfig(
    model='YOUR_MODEL',
    api_url='OPENAI_API_COMPAT_URL',
    api_key='EMPTY_TOKEN',
    datasets=['sciq'],
    limit=10,  # Remove this line for formal evaluation
)

run_task(task_cfg=task_cfg)