---
dataset_info:
  features:
  - name: id
    dtype: string
  - name: submitter
    dtype: string
  - name: authors
    dtype: string
  - name: title
    dtype: string
  - name: comments
    dtype: string
  - name: journal-ref
    dtype: string
  - name: doi
    dtype: string
  - name: report-no
    dtype: string
  - name: categories
    dtype: string
  - name: license
    dtype: string
  - name: orig_abstract
    dtype: string
  - name: versions
    list:
    - name: created
      dtype: string
    - name: version
      dtype: string
  - name: update_date
    dtype: string
  - name: authors_parsed
    sequence:
      sequence: string
  - name: abstract
    dtype: string
  splits:
  - name: train
    num_bytes: 295844767.0072125
    num_examples: 137136
  - name: test
    num_bytes: 63396848.15103951
    num_examples: 29387
  - name: validation
    num_bytes: 63394690.841747954
    num_examples: 29386
  download_size: 236976269
  dataset_size: 422636305.99999994
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: test
    path: data/test-*
  - split: validation
    path: data/validation-*
---


# Dataset Card for hep-ph_gr-qc_primary Dataset

## Dataset Description

- **Homepage:** [Kaggle arXiv Dataset Homepage](https://www.kaggle.com/Cornell-University/arxiv)
- **Repository:** [hepthLlama](https://github.com/Paul-Richmond/hepthLlama)
- **Paper:** [tbd](tbd)
- **Point of Contact:** [Paul Richmond](mailto:p.richmond@qmul.ac.uk)

### Dataset Summary
This dataset contains metadata included in arXiv submissions.

## Dataset Structure

### Languages

The text in the `abstract` field of the dataset is in English, however there may be examples 
where the abstract also contains a translation into another language. 

## Dataset Creation

### Curation Rationale
The starting point was to load v193 of the Kaggle arXiv Dataset which includes arXiv submissions upto 23rd August 2024.
The arXiv dataset contains the following data fields:
- `id`: ArXiv ID (can be used to access the paper)
- `submitter`: Who submitted the paper
- `authors`: Authors of the paper
- `title`: Title of the paper
- `comments`: Additional info, such as number of pages and figures
- `journal-ref`: Information about the journal the paper was published in
- `doi`: [Digital Object Identifier](https://www.doi.org)
- `report-no`: Report Number
- `abstract`: The abstract of the paper
- `categories`: Categories / tags in the ArXiv system

To arrive at the hep-ph_gr-qc_primary dataset, the full arXiv data 
was filtered so that only `categories` which included 'hep-ph' or 'gr-qc' were retained.
This resulted in papers that were either primarily classified as 'hep-ph' or 'gr-qc' or appeared cross-listed.
For this dataset, the decision was made to focus only on papers primarily classified as either 'hep-ph' or 'gr-qc'. 
This meant taking only those abstracts where the first characters in `categories` were either 'hep-ph' or 'gr-qc' 
(see [here](https://info.arxiv.org/help/arxiv_identifier_for_services.html#indications-of-classification) for more details).

We also dropped entries whose `abstract` or `comments` contained the word 'Withdrawn' or 'withdrawn' and we removed the five records which appear in the repo `LLMsForHepth/arxiv_hepth_first_overfit`.

In addition, we have cleaned the data appearing in `abstract` by first replacing all occurences of '\n' with a whitespace and then removing any leading and trailing whitespace.

### Data splits

The dataset is split into a training, validation and test set with split percentages 70%, 15% and 15%. This was done by applying `train_test_split` twice (both with `seed=42`).
The final split sizes are as follows:

| Train | Test | Validation |
|:---:|:---:|:---:|
|137,136 | 29,387| 29,386   |