Edit model card

Model Card of lmqg/mt5-small-frquad-qg-ae

This model is fine-tuned version of google/mt5-small for question generation and answer extraction jointly on the lmqg/qg_frquad (dataset_name: default) via lmqg.

Overview

Usage

from lmqg import TransformersQG

# initialize model
model = TransformersQG(language="fr", model="lmqg/mt5-small-frquad-qg-ae")

# model prediction
question_answer_pairs = model.generate_qa("Créateur » (Maker), lui aussi au singulier, « le Suprême Berger » (The Great Shepherd) ; de l'autre, des réminiscences de la théologie de l'Antiquité : le tonnerre, voix de Jupiter, « Et souvent ta voix gronde en un tonnerre terrifiant », etc.")
  • With transformers
from transformers import pipeline

pipe = pipeline("text2text-generation", "lmqg/mt5-small-frquad-qg-ae")

# answer extraction
answer = pipe("generate question: Créateur » (Maker), lui aussi au singulier, « <hl> le Suprême Berger <hl> » (The Great Shepherd) ; de l'autre, des réminiscences de la théologie de l'Antiquité : le tonnerre, voix de Jupiter, « Et souvent ta voix gronde en un tonnerre terrifiant », etc.")

# question generation
question = pipe("extract answers: Pourtant, la strophe spensérienne, utilisée cinq fois avant que ne commence le chœur, constitue en soi un vecteur dont les répétitions structurelles, selon Ricks, relèvent du pur lyrisme tout en constituant une menace potentielle. Après les huit sages pentamètres iambiques, l'alexandrin final <hl> permet une pause <hl>, « véritable illusion d'optique » qu'accentuent les nombreuses expressions archaïsantes telles que did swoon, did seem, did go, did receive, did make, qui doublent le prétérit en un temps composé et paraissent à la fois « très précautionneuses et très peu pressées ».")

Evaluation

Score Type Dataset
BERTScore 79.9 default lmqg/qg_frquad
Bleu_1 27.6 default lmqg/qg_frquad
Bleu_2 16.31 default lmqg/qg_frquad
Bleu_3 11 default lmqg/qg_frquad
Bleu_4 7.75 default lmqg/qg_frquad
METEOR 17.62 default lmqg/qg_frquad
MoverScore 56.44 default lmqg/qg_frquad
ROUGE_L 28.06 default lmqg/qg_frquad
Score Type Dataset
QAAlignedF1Score (BERTScore) 79.7 default lmqg/qg_frquad
QAAlignedF1Score (MoverScore) 54.22 default lmqg/qg_frquad
QAAlignedPrecision (BERTScore) 77.29 default lmqg/qg_frquad
QAAlignedPrecision (MoverScore) 52.84 default lmqg/qg_frquad
QAAlignedRecall (BERTScore) 82.36 default lmqg/qg_frquad
QAAlignedRecall (MoverScore) 55.76 default lmqg/qg_frquad
Score Type Dataset
AnswerExactMatch 46.96 default lmqg/qg_frquad
AnswerF1Score 67.44 default lmqg/qg_frquad
BERTScore 87.84 default lmqg/qg_frquad
Bleu_1 40.67 default lmqg/qg_frquad
Bleu_2 35.92 default lmqg/qg_frquad
Bleu_3 32.1 default lmqg/qg_frquad
Bleu_4 28.71 default lmqg/qg_frquad
METEOR 37.9 default lmqg/qg_frquad
MoverScore 76.45 default lmqg/qg_frquad
ROUGE_L 43.93 default lmqg/qg_frquad

Training hyperparameters

The following hyperparameters were used during fine-tuning:

  • dataset_path: lmqg/qg_frquad
  • dataset_name: default
  • input_types: ['paragraph_answer', 'paragraph_sentence']
  • output_types: ['question', 'answer']
  • prefix_types: ['qg', 'ae']
  • model: google/mt5-small
  • max_length: 512
  • max_length_output: 32
  • epoch: 18
  • batch: 64
  • lr: 0.0005
  • fp16: False
  • random_seed: 1
  • gradient_accumulation_steps: 1
  • label_smoothing: 0.15

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
23
Inference Examples
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social visibility and check back later, or deploy to Inference Endpoints (dedicated) instead.

Evaluation results

  • BLEU4 (Question Generation) on lmqg/qg_frquad
    self-reported
    7.750
  • ROUGE-L (Question Generation) on lmqg/qg_frquad
    self-reported
    28.060
  • METEOR (Question Generation) on lmqg/qg_frquad
    self-reported
    17.620
  • BERTScore (Question Generation) on lmqg/qg_frquad
    self-reported
    79.900
  • MoverScore (Question Generation) on lmqg/qg_frquad
    self-reported
    56.440
  • QAAlignedF1Score-BERTScore (Question & Answer Generation (with Gold Answer)) on lmqg/qg_frquad
    self-reported
    79.700
  • QAAlignedRecall-BERTScore (Question & Answer Generation (with Gold Answer)) on lmqg/qg_frquad
    self-reported
    82.360
  • QAAlignedPrecision-BERTScore (Question & Answer Generation (with Gold Answer)) on lmqg/qg_frquad
    self-reported
    77.290
  • QAAlignedF1Score-MoverScore (Question & Answer Generation (with Gold Answer)) on lmqg/qg_frquad
    self-reported
    54.220
  • QAAlignedRecall-MoverScore (Question & Answer Generation (with Gold Answer)) on lmqg/qg_frquad
    self-reported
    55.760