Model Card of lmqg/mbart-large-cc25-itquad-qg-ae
This model is fine-tuned version of facebook/mbart-large-cc25 for question generation and answer extraction jointly on the lmqg/qg_itquad (dataset_name: default) via lmqg
.
Overview
- Language model: facebook/mbart-large-cc25
- Language: it
- Training data: lmqg/qg_itquad (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="it", model="lmqg/mbart-large-cc25-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/mbart-large-cc25-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
- Metric (Question Generation): raw metric file
Score | Type | Dataset | |
---|---|---|---|
BERTScore | 79.29 | default | lmqg/qg_itquad |
Bleu_1 | 22.03 | default | lmqg/qg_itquad |
Bleu_2 | 14.31 | default | lmqg/qg_itquad |
Bleu_3 | 9.9 | default | lmqg/qg_itquad |
Bleu_4 | 7.06 | default | lmqg/qg_itquad |
METEOR | 16.86 | default | lmqg/qg_itquad |
MoverScore | 55.92 | default | lmqg/qg_itquad |
ROUGE_L | 20.15 | default | lmqg/qg_itquad |
- Metric (Question & Answer Generation): raw metric file
Score | Type | Dataset | |
---|---|---|---|
QAAlignedF1Score (BERTScore) | 82.65 | default | lmqg/qg_itquad |
QAAlignedF1Score (MoverScore) | 56.14 | default | lmqg/qg_itquad |
QAAlignedPrecision (BERTScore) | 81.06 | default | lmqg/qg_itquad |
QAAlignedPrecision (MoverScore) | 55.22 | default | lmqg/qg_itquad |
QAAlignedRecall (BERTScore) | 84.34 | default | lmqg/qg_itquad |
QAAlignedRecall (MoverScore) | 57.13 | default | lmqg/qg_itquad |
- Metric (Answer Extraction): raw metric file
Score | Type | Dataset | |
---|---|---|---|
AnswerExactMatch | 63.88 | default | lmqg/qg_itquad |
AnswerF1Score | 76.59 | default | lmqg/qg_itquad |
BERTScore | 90.63 | default | lmqg/qg_itquad |
Bleu_1 | 33.66 | default | lmqg/qg_itquad |
Bleu_2 | 27.96 | default | lmqg/qg_itquad |
Bleu_3 | 23.79 | default | lmqg/qg_itquad |
Bleu_4 | 20.21 | default | lmqg/qg_itquad |
METEOR | 44.48 | default | lmqg/qg_itquad |
MoverScore | 83.05 | default | lmqg/qg_itquad |
ROUGE_L | 46.51 | 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: facebook/mbart-large-cc25
- max_length: 512
- max_length_output: 32
- epoch: 8
- batch: 2
- lr: 0.0001
- fp16: False
- random_seed: 1
- gradient_accumulation_steps: 32
- 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
- 7
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.
Dataset used to train research-backup/mbart-large-cc25-itquad-qg-ae
Evaluation results
- BLEU4 (Question Generation) on lmqg/qg_itquadself-reported7.060
- ROUGE-L (Question Generation) on lmqg/qg_itquadself-reported20.150
- METEOR (Question Generation) on lmqg/qg_itquadself-reported16.860
- BERTScore (Question Generation) on lmqg/qg_itquadself-reported79.290
- MoverScore (Question Generation) on lmqg/qg_itquadself-reported55.920
- QAAlignedF1Score-BERTScore (Question & Answer Generation (with Gold Answer)) on lmqg/qg_itquadself-reported82.650
- QAAlignedRecall-BERTScore (Question & Answer Generation (with Gold Answer)) on lmqg/qg_itquadself-reported84.340
- QAAlignedPrecision-BERTScore (Question & Answer Generation (with Gold Answer)) on lmqg/qg_itquadself-reported81.060
- QAAlignedF1Score-MoverScore (Question & Answer Generation (with Gold Answer)) on lmqg/qg_itquadself-reported56.140
- QAAlignedRecall-MoverScore (Question & Answer Generation (with Gold Answer)) on lmqg/qg_itquadself-reported57.130