Edit model card

Model Card of research-backup/bart-large-squad-qg-default

This model is fine-tuned version of facebook/bart-large for question generation task on the lmqg/qg_squad (dataset_name: default) via lmqg. This model is fine-tuned without parameter search (default configuration is taken from ERNIE-GEN).

Overview

Usage

from lmqg import TransformersQG

# initialize model
model = TransformersQG(language="en", model="research-backup/bart-large-squad-qg-default")

# model prediction
questions = model.generate_q(list_context="William Turner was an English painter who specialised in watercolour landscapes", list_answer="William Turner")
  • With transformers
from transformers import pipeline

pipe = pipeline("text2text-generation", "research-backup/bart-large-squad-qg-default")
output = pipe("<hl> Beyonce <hl> further expanded her acting career, starring as blues singer Etta James in the 2008 musical biopic, Cadillac Records.")

Evaluation

Score Type Dataset
BERTScore 90.95 default lmqg/qg_squad
Bleu_1 56.25 default lmqg/qg_squad
Bleu_2 40.27 default lmqg/qg_squad
Bleu_3 30.71 default lmqg/qg_squad
Bleu_4 23.94 default lmqg/qg_squad
METEOR 25.91 default lmqg/qg_squad
MoverScore 64.42 default lmqg/qg_squad
ROUGE_L 52.2 default lmqg/qg_squad

Training hyperparameters

The following hyperparameters were used during fine-tuning:

  • dataset_path: lmqg/qg_squad
  • dataset_name: default
  • input_types: ['paragraph_answer']
  • output_types: ['question']
  • prefix_types: None
  • model: facebook/bart-large
  • max_length: 512
  • max_length_output: 32
  • epoch: 10
  • batch: 8
  • lr: 1.25e-05
  • fp16: False
  • random_seed: 1
  • gradient_accumulation_steps: 4
  • label_smoothing: 0.1

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",
}
Downloads last month
21

Dataset used to train research-backup/bart-large-squad-qg-default

Evaluation results