Verified71%Harbor: pendingKnowledge & Reasoning

News QA

Question answering using news articles.

How to run

ltb · news-qa
$ npm i -g lettokenburn-cli
$ ltb bench pull news-qa

External harness. pip install crfm-helm && helm-run --run-entries newsqa:model=<model> --suite my-suite --max-eval-instances 100 (HELM scenario 'newsqa'; some scenarios need extra args) https://github.com/stanford-crfm/helm

Runs happen on your machine with your own subscription; only the score summary is pushed. CLI docs

Scores

8 sourced entries
HarnessSourceConfidenceLink
OpenAI/o3-2025-01-31-high92.5N/AIndependentmediumJan 30, 2025arxiv.org
OpenAI/o4-mini-2025-05-01-high88.2N/AIndependentmediumJun 13, 2021arxiv.org
OpenAI/o1-2024-12-17-high81.4N/AIndependentmediumJun 19, 2022arxiv.org
DeepSeek R180.6N/AIndependentmediumJan 22, 2025arxiv.org
Anthropic/claude-3-7-sonnet74.2N/AIndependentmediumJun 24, 2024www-cdn.anthropic.com
xAI/grok-3-121270.6N/AIndependentmediumFeb 15, 2022arxiv.org
GPT-4o70.2N/AIndependentmediumOct 3, 2024arxiv.org
Gemini 2.5 Pro68.8N/AIndependentmediumMar 8, 2024arxiv.org

Every number links to where it was published. Low-confidence rows are muted and need per-cell verification.

Reproduced by the community

All runs
Loading community runs…

About

The NewsQA dataset is from the paper: https://arxiv.org/abs/1611.09830

Original repository can be found at: https://github.com/Maluuba/newsqa

Note: The training dataset cannot be directly shared due to copyright issues, and needs to be downloaded by following the instructions in the repo above. These instructions are duplicated here for convenience.

  1. Clone the repo (https://github.com/Maluuba/newsqa)
  2. Download the data from (https://msropendata.com/datasets/939b1042-6402-4697-9c15-7a28de7e1321).

You need to create a login account to download this data.

  1. Download the CNN stories tar file from "https://cs.nyu.edu/~kcho/DMQA/"
  2. Create the conda environment using the command (conda create --name newsqa python=2.7 "pandas>=0.19.2")
  3. Install the requirements (conda activate newsqa && pip install --requirement requirements.txt)

This should result in the creation of the file (combined-newsqa-data-v1.json) in the repo which is used in this scenario.

NewsQA is a QA dataset containing 12,744 stories, and over 119,633 question-answer pairs. There are 92549 training qa pairs, 5166 qas in the dev set, and 5126 in the test set. Particularly, given the a news article from CNN, the goal is answer questions with answers consisting of spans of text from the corresponding articles. All of the questions and answers are written by crowd sourced human annotators. For more details, see https://arxiv.org/abs/1611.09830.

More concretely, we prompt models using the following format

Passage: <news article> Question: <question> Answer:

Note: Some of the questions do not have an answer in the context so the model needs to answer "No Answer". While this behavior might be tricky to learn in the few-shot setting, we still include these examples in the scenario.

Using an example from the training dataset, we have:

NEW DELHI, India (CNN) -- A high court in northern India on Friday acquitted a wealthy businessman
facing the death sentence for the killing of a teen in a case dubbed 'the house of horrors.'
Moninder Singh Pandher was sentenced to death by a lower court in February...
Question: Who was sentenced to death in February?
Answer:

References

['Moninder Singh Pandher']