Edit model card

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

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

Overview

Usage

from lmqg import TransformersQG

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

# model prediction
question_answer_pairs = model.generate_qa("Dopo il 1971 , l' OPEC ha tardato ad adeguare i prezzi per riflettere tale deprezzamento.")
  • With transformers
from transformers import pipeline

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

# answer extraction
answer = pipe("generate question: <hl> Dopo il 1971 <hl> , l' OPEC ha tardato ad adeguare i prezzi per riflettere tale deprezzamento.")

# question generation
question = pipe("extract answers: <hl> Il 6 ottobre 1973 , la Siria e l' Egitto, con il sostegno di altre nazioni arabe, lanciarono un attacco a sorpresa su Israele, su Yom Kippur. <hl> Questo rinnovo delle ostilità nel conflitto arabo-israeliano ha liberato la pressione economica sottostante sui prezzi del petrolio. All' epoca, l' Iran era il secondo esportatore mondiale di petrolio e un vicino alleato degli Stati Uniti. Settimane più tardi, lo scià d' Iran ha detto in un' intervista: Naturalmente[il prezzo del petrolio] sta andando a salire Certamente! E come! Avete[Paesi occidentali] aumentato il prezzo del grano che ci vendete del 300 per cento, e lo stesso per zucchero e cemento.")

Evaluation

Score Type Dataset
BERTScore 80.61 default lmqg/qg_itquad
Bleu_1 22.53 default lmqg/qg_itquad
Bleu_2 14.75 default lmqg/qg_itquad
Bleu_3 10.19 default lmqg/qg_itquad
Bleu_4 7.25 default lmqg/qg_itquad
METEOR 17.5 default lmqg/qg_itquad
MoverScore 56.63 default lmqg/qg_itquad
ROUGE_L 21.84 default lmqg/qg_itquad
Score Type Dataset
QAAlignedF1Score (BERTScore) 81.81 default lmqg/qg_itquad
QAAlignedF1Score (MoverScore) 56.02 default lmqg/qg_itquad
QAAlignedPrecision (BERTScore) 81.17 default lmqg/qg_itquad
QAAlignedPrecision (MoverScore) 55.76 default lmqg/qg_itquad
QAAlignedRecall (BERTScore) 82.51 default lmqg/qg_itquad
QAAlignedRecall (MoverScore) 56.32 default lmqg/qg_itquad
Score Type Dataset
AnswerExactMatch 57.85 default lmqg/qg_itquad
AnswerF1Score 72.09 default lmqg/qg_itquad
BERTScore 90.24 default lmqg/qg_itquad
Bleu_1 39.33 default lmqg/qg_itquad
Bleu_2 33.64 default lmqg/qg_itquad
Bleu_3 29.59 default lmqg/qg_itquad
Bleu_4 26.01 default lmqg/qg_itquad
METEOR 42.68 default lmqg/qg_itquad
MoverScore 81.17 default lmqg/qg_itquad
ROUGE_L 45.15 default lmqg/qg_itquad

Training hyperparameters

The following hyperparameters were used during fine-tuning:

  • dataset_path: lmqg/qg_itquad
  • 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: 13
  • batch: 16
  • lr: 0.001
  • fp16: False
  • random_seed: 1
  • gradient_accumulation_steps: 4
  • 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
15

Dataset used to train lmqg/mt5-small-itquad-qg-ae

Evaluation results

  • BLEU4 (Question Generation) on lmqg/qg_itquad
    self-reported
    7.250
  • ROUGE-L (Question Generation) on lmqg/qg_itquad
    self-reported
    21.840
  • METEOR (Question Generation) on lmqg/qg_itquad
    self-reported
    17.500
  • BERTScore (Question Generation) on lmqg/qg_itquad
    self-reported
    80.610
  • MoverScore (Question Generation) on lmqg/qg_itquad
    self-reported
    56.630
  • QAAlignedF1Score-BERTScore (Question & Answer Generation (with Gold Answer)) on lmqg/qg_itquad
    self-reported
    81.810
  • QAAlignedRecall-BERTScore (Question & Answer Generation (with Gold Answer)) on lmqg/qg_itquad
    self-reported
    82.510
  • QAAlignedPrecision-BERTScore (Question & Answer Generation (with Gold Answer)) on lmqg/qg_itquad
    self-reported
    81.170
  • QAAlignedF1Score-MoverScore (Question & Answer Generation (with Gold Answer)) on lmqg/qg_itquad
    self-reported
    56.020
  • QAAlignedRecall-MoverScore (Question & Answer Generation (with Gold Answer)) on lmqg/qg_itquad
    self-reported
    56.320
  • QAAlignedPrecision-MoverScore (Question & Answer Generation (with Gold Answer)) on lmqg/qg_itquad
    self-reported
    55.760
  • BLEU4 (Answer Extraction) on lmqg/qg_itquad
    self-reported
    26.010
  • ROUGE-L (Answer Extraction) on lmqg/qg_itquad
    self-reported
    45.150
  • METEOR (Answer Extraction) on lmqg/qg_itquad
    self-reported
    42.680
  • BERTScore (Answer Extraction) on lmqg/qg_itquad
    self-reported
    90.240
  • MoverScore (Answer Extraction) on lmqg/qg_itquad
    self-reported
    81.170
  • AnswerF1Score (Answer Extraction) on lmqg/qg_itquad
    self-reported
    72.090
  • AnswerExactMatch (Answer Extraction) on lmqg/qg_itquad
    self-reported
    57.850