Model Card of lmqg/t5-base-tweetqa-qa

This model is fine-tuned version of t5-base for question answering task on the lmqg/qg_tweetqa (dataset_name: default) via lmqg.

Overview

Usage

from lmqg import TransformersQG

# initialize model
model = TransformersQG(language="en", model="lmqg/t5-base-tweetqa-qa")

# model prediction
answers = model.answer_q(list_question="What is a person called is practicing heresy?", list_context=" Heresy is any provocative belief or theory that is strongly at variance with established beliefs or customs. A heretic is a proponent of such claims or beliefs. Heresy is distinct from both apostasy, which is the explicit renunciation of one's religion, principles or cause, and blasphemy, which is an impious utterance or action concerning God or sacred things.")
  • With transformers
from transformers import pipeline

pipe = pipeline("text2text-generation", "lmqg/t5-base-tweetqa-qa")
output = pipe("question: What is a person called is practicing heresy?, context: Heresy is any provocative belief or theory that is strongly at variance with established beliefs or customs. A heretic is a proponent of such claims or beliefs. Heresy is distinct from both apostasy, which is the explicit renunciation of one's religion, principles or cause, and blasphemy, which is an impious utterance or action concerning God or sacred things.")

Evaluation

Score Type Dataset
AnswerExactMatch 52.29 default lmqg/qg_tweetqa
AnswerF1Score 69.4 default lmqg/qg_tweetqa
BERTScore 94.58 default lmqg/qg_tweetqa
Bleu_1 57.07 default lmqg/qg_tweetqa
Bleu_2 48.17 default lmqg/qg_tweetqa
Bleu_3 39.78 default lmqg/qg_tweetqa
Bleu_4 33.32 default lmqg/qg_tweetqa
METEOR 35.43 default lmqg/qg_tweetqa
MoverScore 80.07 default lmqg/qg_tweetqa
ROUGE_L 62.24 default lmqg/qg_tweetqa

Training hyperparameters

The following hyperparameters were used during fine-tuning:

  • dataset_path: lmqg/qg_tweetqa
  • dataset_name: default
  • input_types: ['paragraph_question']
  • output_types: ['answer']
  • prefix_types: None
  • model: t5-base
  • max_length: 512
  • max_length_output: 32
  • epoch: 10
  • batch: 32
  • lr: 0.0001
  • fp16: False
  • random_seed: 1
  • gradient_accumulation_steps: 2
  • 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
9
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.

Dataset used to train lmqg/t5-base-tweetqa-qa

Evaluation results

  • BLEU4 (Question Answering) on lmqg/qg_tweetqa
    self-reported
    33.320
  • ROUGE-L (Question Answering) on lmqg/qg_tweetqa
    self-reported
    62.240
  • METEOR (Question Answering) on lmqg/qg_tweetqa
    self-reported
    35.430
  • BERTScore (Question Answering) on lmqg/qg_tweetqa
    self-reported
    94.580
  • MoverScore (Question Answering) on lmqg/qg_tweetqa
    self-reported
    80.070
  • AnswerF1Score (Question Answering) on lmqg/qg_tweetqa
    self-reported
    69.400
  • AnswerExactMatch (Question Answering) on lmqg/qg_tweetqa
    self-reported
    52.290