Model Card of vocabtrimmer/mt5-small-trimmed-ru-120000-ruquad-qg
This model is fine-tuned version of vocabtrimmer/mt5-small-trimmed-ru-120000 for question generation task on the lmqg/qg_ruquad (dataset_name: default) via lmqg
.
Overview
- Language model: vocabtrimmer/mt5-small-trimmed-ru-120000
- Language: ru
- Training data: lmqg/qg_ruquad (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="ru", model="vocabtrimmer/mt5-small-trimmed-ru-120000-ruquad-qg")
# model prediction
questions = model.generate_q(list_context="Нелишним будет отметить, что, развивая это направление, Д. И. Менделеев, поначалу априорно выдвинув идею о температуре, при которой высота мениска будет нулевой, в мае 1860 года провёл серию опытов.", list_answer="в мае 1860 года")
- With
transformers
from transformers import pipeline
pipe = pipeline("text2text-generation", "vocabtrimmer/mt5-small-trimmed-ru-120000-ruquad-qg")
output = pipe("Нелишним будет отметить, что, развивая это направление, Д. И. Менделеев, поначалу априорно выдвинув идею о температуре, при которой высота мениска будет нулевой, <hl> в мае 1860 года <hl> провёл серию опытов.")
Evaluation
- Metric (Question Generation): raw metric file
Score | Type | Dataset | |
---|---|---|---|
BERTScore | 86.01 | default | lmqg/qg_ruquad |
Bleu_1 | 33.64 | default | lmqg/qg_ruquad |
Bleu_2 | 26.89 | default | lmqg/qg_ruquad |
Bleu_3 | 21.94 | default | lmqg/qg_ruquad |
Bleu_4 | 18.11 | default | lmqg/qg_ruquad |
METEOR | 28.94 | default | lmqg/qg_ruquad |
MoverScore | 64.61 | default | lmqg/qg_ruquad |
ROUGE_L | 33.73 | default | lmqg/qg_ruquad |
Training hyperparameters
The following hyperparameters were used during fine-tuning:
- dataset_path: lmqg/qg_ruquad
- dataset_name: default
- input_types: paragraph_answer
- output_types: question
- prefix_types: None
- model: vocabtrimmer/mt5-small-trimmed-ru-120000
- max_length: 512
- max_length_output: 32
- epoch: 14
- batch: 16
- lr: 0.0005
- 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",
}