---
dataset_info:
  features:
  - name: query
    dtype: string
  - name: pos
    dtype: string
  - name: neg
    dtype: string
  - name: query_ids
    sequence: int32
  - name: query_attention_mask
    sequence: int8
  - name: positive_ids
    sequence: int32
  - name: positive_attention_mask
    sequence: int8
  - name: negative_ids
    sequence: int32
  - name: negative_attention_mask
    sequence: int8
  splits:
  - name: train
    num_bytes: 2761369465.438713
    num_examples: 1986832
  - name: validation
    num_bytes: 9385561.547585024
    num_examples: 6522
  download_size: 1141983769
  dataset_size: 2770755026.986298
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: validation
    path: data/validation-*
---

This dataset is created to train language models on their retrieval and re-ranking abilities. More detailed information can be found at our publication: [Advancing Vietnamese Information Retrieval with Learning Objective and Benchmark](https://arxiv.org/abs/2503.07470).

Please cite this paper if you want to use it for your work.