Model Card of lmqg/flan-t5-base-squad-qag
This model is fine-tuned version of google/flan-t5-base for question & answer pair generation task on the lmqg/qag_squad (dataset_name: default) via lmqg
.
Overview
- Language model: google/flan-t5-base
- Language: en
- Training data: lmqg/qag_squad (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="en", model="lmqg/flan-t5-base-squad-qag")
# model prediction
question_answer_pairs = model.generate_qa("William Turner was an English painter who specialised in watercolour landscapes")
- With
transformers
from transformers import pipeline
pipe = pipeline("text2text-generation", "lmqg/flan-t5-base-squad-qag")
output = pipe("generate question and answer: Beyonce further expanded her acting career, starring as blues singer Etta James in the 2008 musical biopic, Cadillac Records.")
Evaluation
- Metric (Question & Answer Generation): raw metric file
Score | Type | Dataset | |
---|---|---|---|
QAAlignedF1Score (BERTScore) | 93.04 | default | lmqg/qag_squad |
QAAlignedF1Score (MoverScore) | 65.24 | default | lmqg/qag_squad |
QAAlignedPrecision (BERTScore) | 93.1 | default | lmqg/qag_squad |
QAAlignedPrecision (MoverScore) | 65.91 | default | lmqg/qag_squad |
QAAlignedRecall (BERTScore) | 92.99 | default | lmqg/qag_squad |
QAAlignedRecall (MoverScore) | 64.7 | default | lmqg/qag_squad |
Training hyperparameters
The following hyperparameters were used during fine-tuning:
- dataset_path: lmqg/qag_squad
- dataset_name: default
- input_types: ['paragraph']
- output_types: ['questions_answers']
- prefix_types: ['qag']
- model: google/flan-t5-base
- max_length: 512
- max_length_output: 256
- epoch: 14
- batch: 8
- lr: 0.0001
- fp16: False
- random_seed: 1
- gradient_accumulation_steps: 8
- 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",
}
- Downloads last month
- 21
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 lmqg/flan-t5-base-squad-qag
Evaluation results
- QAAlignedF1Score-BERTScore (Question & Answer Generation) on lmqg/qag_squadself-reported93.040
- QAAlignedRecall-BERTScore (Question & Answer Generation) on lmqg/qag_squadself-reported92.990
- QAAlignedPrecision-BERTScore (Question & Answer Generation) on lmqg/qag_squadself-reported93.100
- QAAlignedF1Score-MoverScore (Question & Answer Generation) on lmqg/qag_squadself-reported65.240
- QAAlignedRecall-MoverScore (Question & Answer Generation) on lmqg/qag_squadself-reported64.700
- QAAlignedPrecision-MoverScore (Question & Answer Generation) on lmqg/qag_squadself-reported65.910