metadata

license: cc-by-4.0
metrics:
  - bleu4
  - meteor
  - rouge-l
  - bertscore
  - moverscore
language: ko
datasets:
  - lmqg/qag_koquad
pipeline_tag: text2text-generation
tags:
  - questions and answers generation
widget:
  - text: >-
      1990년 영화 《 남부군 》에서 단역으로 영화배우 첫 데뷔에 이어 같은 해 KBS 드라마 《지구인》에서 단역으로 출연하였고 이듬해
      MBC 《여명의 눈동자》를 통해 단역으로 출연하였다.
    example_title: Questions & Answers Generation Example 1
model-index:
  - name: lmqg/mbart-large-cc25-koquad-qag
    results:
      - task:
          name: Text2text Generation
          type: text2text-generation
        dataset:
          name: lmqg/qag_koquad
          type: default
          args: default
        metrics:
          - name: BLEU4 (Question & Answer Generation)
            type: bleu4_question_answer_generation
            value: 2.31
          - name: ROUGE-L (Question & Answer Generation)
            type: rouge_l_question_answer_generation
            value: 14.7
          - name: METEOR (Question & Answer Generation)
            type: meteor_question_answer_generation
            value: 23.98
          - name: BERTScore (Question & Answer Generation)
            type: bertscore_question_answer_generation
            value: 67.01
          - name: MoverScore (Question & Answer Generation)
            type: moverscore_question_answer_generation
            value: 63.18
          - name: QAAlignedF1Score-BERTScore (Question & Answer Generation)
            type: qa_aligned_f1_score_bertscore_question_answer_generation
            value: 80.71
          - name: QAAlignedRecall-BERTScore (Question & Answer Generation)
            type: qa_aligned_recall_bertscore_question_answer_generation
            value: 83.05
          - name: QAAlignedPrecision-BERTScore (Question & Answer Generation)
            type: qa_aligned_precision_bertscore_question_answer_generation
            value: 78.58
          - name: QAAlignedF1Score-MoverScore (Question & Answer Generation)
            type: qa_aligned_f1_score_moverscore_question_answer_generation
            value: 81.75
          - name: QAAlignedRecall-MoverScore (Question & Answer Generation)
            type: qa_aligned_recall_moverscore_question_answer_generation
            value: 84.86
          - name: QAAlignedPrecision-MoverScore (Question & Answer Generation)
            type: qa_aligned_precision_moverscore_question_answer_generation
            value: 79.01

Model Card of `lmqg/mbart-large-cc25-koquad-qag`

This model is fine-tuned version of facebook/mbart-large-cc25 for question & answer pair generation task on the lmqg/qag_koquad (dataset_name: default) via lmqg.

Overview

Language model: facebook/mbart-large-cc25
Language: ko
Training data: lmqg/qag_koquad (default)
Online Demo: https://autoqg.net/
Repository: https://github.com/asahi417/lm-question-generation
Paper: https://arxiv.org/abs/2210.03992

Usage

With lmqg

from lmqg import TransformersQG

# initialize model
model = TransformersQG(language="ko", model="lmqg/mbart-large-cc25-koquad-qag")

# model prediction
question_answer_pairs = model.generate_qa("1990년 영화 《 남부군 》에서 단역으로 영화배우 첫 데뷔에 이어 같은 해 KBS 드라마 《지구인》에서 단역으로 출연하였고 이듬해 MBC 《여명의 눈동자》를 통해 단역으로 출연하였다.")

With transformers

from transformers import pipeline

pipe = pipeline("text2text-generation", "lmqg/mbart-large-cc25-koquad-qag")
output = pipe("1990년 영화 《 남부군 》에서 단역으로 영화배우 첫 데뷔에 이어 같은 해 KBS 드라마 《지구인》에서 단역으로 출연하였고 이듬해 MBC 《여명의 눈동자》를 통해 단역으로 출연하였다.")

Evaluation

Metric (Question & Answer Generation): raw metric file

	Score	Type	Dataset
BERTScore	67.01	default	lmqg/qag_koquad
Bleu_1	8.53	default	lmqg/qag_koquad
Bleu_2	5.29	default	lmqg/qag_koquad
Bleu_3	3.45	default	lmqg/qag_koquad
Bleu_4	2.31	default	lmqg/qag_koquad
METEOR	23.98	default	lmqg/qag_koquad
MoverScore	63.18	default	lmqg/qag_koquad
QAAlignedF1Score (BERTScore)	80.71	default	lmqg/qag_koquad
QAAlignedF1Score (MoverScore)	81.75	default	lmqg/qag_koquad
QAAlignedPrecision (BERTScore)	78.58	default	lmqg/qag_koquad
QAAlignedPrecision (MoverScore)	79.01	default	lmqg/qag_koquad
QAAlignedRecall (BERTScore)	83.05	default	lmqg/qag_koquad
QAAlignedRecall (MoverScore)	84.86	default	lmqg/qag_koquad
ROUGE_L	14.7	default	lmqg/qag_koquad

Training hyperparameters

The following hyperparameters were used during fine-tuning:

dataset_path: lmqg/qag_koquad
dataset_name: default
input_types: ['paragraph']
output_types: ['questions_answers']
prefix_types: None
model: facebook/mbart-large-cc25
max_length: 512
max_length_output: 256
epoch: 11
batch: 2
lr: 0.0001
fp16: False
random_seed: 1
gradient_accumulation_steps: 32
label_smoothing: 0.0

The full configuration can be found at fine-tuning config file.

Citation

@inproceedings{ushio-etal-2022-generative,
    title = "{G}enerative {L}anguage {M}odels for {P}aragraph-{L}evel {Q}uestion {G}eneration",
    author = "Ushio, Asahi  and
        Alva-Manchego, Fernando  and
        Camacho-Collados, Jose",
    booktitle = "Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing",
    month = dec,
    year = "2022",
    address = "Abu Dhabi, U.A.E.",
    publisher = "Association for Computational Linguistics",
}