--- license: cc-by-4.0 metrics: - bleu4 - meteor - rouge-l - bertscore - moverscore language: es datasets: - lmqg/qg_esquad pipeline_tag: text2text-generation tags: - question generation - answer extraction widget: - text: "generate question: del Ministerio de Desarrollo Urbano , Gobierno de la India." example_title: "Question Generation Example 1" - text: "generate question: a noviembre , que es también la estación lluviosa." example_title: "Question Generation Example 2" - text: "generate question: como el gobierno de Abbott que asumió el cargo el 18 de septiembre de 2013." example_title: "Question Generation Example 3" - text: "extract answers: En la diáspora somalí, múltiples eventos islámicos de recaudación de fondos se llevan a cabo cada año en ciudades como Birmingham, Londres, Toronto y Minneapolis, donde los académicos y profesionales somalíes dan conferencias y responden preguntas de la audiencia. El propósito de estos eventos es recaudar dinero para nuevas escuelas o universidades en Somalia, para ayudar a los somalíes que han sufrido como consecuencia de inundaciones y / o sequías, o para reunir fondos para la creación de nuevas mezquitas como." example_title: "Answer Extraction Example 1" - text: "extract answers: Los estudiosos y los histori a dores están divididos en cuanto a qué evento señala el final de la era helenística. El período helenístico se puede ver que termina con la conquista final del corazón griego por Roma en 146 a. C. tras la guerra aquea, con la derrota final del reino ptolemaico en la batalla de Actium en 31 a. Helenístico se distingue de helénico en que el primero abarca toda la esfera de influencia griega antigua directa, mientras que el segundo se refiere a la propia Grecia." example_title: "Answer Extraction Example 2" model-index: - name: lmqg/mt5-base-esquad-qg-ae results: - task: name: Text2text Generation type: text2text-generation dataset: name: lmqg/qg_esquad type: default args: default metrics: - name: BLEU4 (Question Generation) type: bleu4_question_generation value: 9.62 - name: ROUGE-L (Question Generation) type: rouge_l_question_generation value: 24.82 - name: METEOR (Question Generation) type: meteor_question_generation value: 23.11 - name: BERTScore (Question Generation) type: bertscore_question_generation value: 83.97 - name: MoverScore (Question Generation) type: moverscore_question_generation value: 59.15 - name: BLEU4 (Question & Answer Generation (with Gold Answer)) type: bleu4_question_answer_generation_with_gold_answer value: 1.78 - name: ROUGE-L (Question & Answer Generation (with Gold Answer)) type: rouge_l_question_answer_generation_with_gold_answer value: 15.15 - name: METEOR (Question & Answer Generation (with Gold Answer)) type: meteor_question_answer_generation_with_gold_answer value: 22.16 - name: BERTScore (Question & Answer Generation (with Gold Answer)) type: bertscore_question_answer_generation_with_gold_answer value: 69.46 - name: MoverScore (Question & Answer Generation (with Gold Answer)) type: moverscore_question_answer_generation_with_gold_answer value: 51.69 - name: QAAlignedF1Score-BERTScore (Question & Answer Generation (with Gold Answer)) type: qa_aligned_f1_score_bertscore_question_answer_generation_with_gold_answer value: 79.67 - name: QAAlignedRecall-BERTScore (Question & Answer Generation (with Gold Answer)) type: qa_aligned_recall_bertscore_question_answer_generation_with_gold_answer value: 82.44 - name: QAAlignedPrecision-BERTScore (Question & Answer Generation (with Gold Answer)) type: qa_aligned_precision_bertscore_question_answer_generation_with_gold_answer value: 77.14 - name: QAAlignedF1Score-MoverScore (Question & Answer Generation (with Gold Answer)) type: qa_aligned_f1_score_moverscore_question_answer_generation_with_gold_answer value: 54.82 - name: QAAlignedRecall-MoverScore (Question & Answer Generation (with Gold Answer)) type: qa_aligned_recall_moverscore_question_answer_generation_with_gold_answer value: 56.56 - name: QAAlignedPrecision-MoverScore (Question & Answer Generation (with Gold Answer)) type: qa_aligned_precision_moverscore_question_answer_generation_with_gold_answer value: 53.27 - name: BLEU4 (Answer Extraction) type: bleu4_answer_extraction value: 25.75 - name: ROUGE-L (Answer Extraction) type: rouge_l_answer_extraction value: 49.61 - name: METEOR (Answer Extraction) type: meteor_answer_extraction value: 43.74 - name: BERTScore (Answer Extraction) type: bertscore_answer_extraction value: 90.04 - name: MoverScore (Answer Extraction) type: moverscore_answer_extraction value: 80.94 - name: AnswerF1Score (Answer Extraction) type: answer_f1_score__answer_extraction value: 75.33 - name: AnswerExactMatch (Answer Extraction) type: answer_exact_match_answer_extraction value: 57.98 --- # Model Card of `lmqg/mt5-base-esquad-qg-ae` This model is fine-tuned version of [google/mt5-base](https://huggingface.co/google/mt5-base) for question generation and answer extraction jointly on the [lmqg/qg_esquad](https://huggingface.co/datasets/lmqg/qg_esquad) (dataset_name: default) via [`lmqg`](https://github.com/asahi417/lm-question-generation). ### Overview - **Language model:** [google/mt5-base](https://huggingface.co/google/mt5-base) - **Language:** es - **Training data:** [lmqg/qg_esquad](https://huggingface.co/datasets/lmqg/qg_esquad) (default) - **Online Demo:** [https://autoqg.net/](https://autoqg.net/) - **Repository:** [https://github.com/asahi417/lm-question-generation](https://github.com/asahi417/lm-question-generation) - **Paper:** [https://arxiv.org/abs/2210.03992](https://arxiv.org/abs/2210.03992) ### Usage - With [`lmqg`](https://github.com/asahi417/lm-question-generation#lmqg-language-model-for-question-generation-) ```python from lmqg import TransformersQG # initialize model model = TransformersQG(language="es", model="lmqg/mt5-base-esquad-qg-ae") # model prediction question_answer_pairs = model.generate_qa("a noviembre , que es también la estación lluviosa.") ``` - With `transformers` ```python from transformers import pipeline pipe = pipeline("text2text-generation", "lmqg/mt5-base-esquad-qg-ae") # answer extraction answer = pipe("generate question: del Ministerio de Desarrollo Urbano , Gobierno de la India.") # question generation question = pipe("extract answers: En la diáspora somalí, múltiples eventos islámicos de recaudación de fondos se llevan a cabo cada año en ciudades como Birmingham, Londres, Toronto y Minneapolis, donde los académicos y profesionales somalíes dan conferencias y responden preguntas de la audiencia. El propósito de estos eventos es recaudar dinero para nuevas escuelas o universidades en Somalia, para ayudar a los somalíes que han sufrido como consecuencia de inundaciones y / o sequías, o para reunir fondos para la creación de nuevas mezquitas como.") ``` ## Evaluation - ***Metric (Question Generation)***: [raw metric file](https://huggingface.co/lmqg/mt5-base-esquad-qg-ae/raw/main/eval/metric.first.sentence.paragraph_answer.question.lmqg_qg_esquad.default.json) | | Score | Type | Dataset | |:-----------|--------:|:--------|:-----------------------------------------------------------------| | BERTScore | 83.97 | default | [lmqg/qg_esquad](https://huggingface.co/datasets/lmqg/qg_esquad) | | Bleu_1 | 25.88 | default | [lmqg/qg_esquad](https://huggingface.co/datasets/lmqg/qg_esquad) | | Bleu_2 | 17.67 | default | [lmqg/qg_esquad](https://huggingface.co/datasets/lmqg/qg_esquad) | | Bleu_3 | 12.84 | default | [lmqg/qg_esquad](https://huggingface.co/datasets/lmqg/qg_esquad) | | Bleu_4 | 9.62 | default | [lmqg/qg_esquad](https://huggingface.co/datasets/lmqg/qg_esquad) | | METEOR | 23.11 | default | [lmqg/qg_esquad](https://huggingface.co/datasets/lmqg/qg_esquad) | | MoverScore | 59.15 | default | [lmqg/qg_esquad](https://huggingface.co/datasets/lmqg/qg_esquad) | | ROUGE_L | 24.82 | default | [lmqg/qg_esquad](https://huggingface.co/datasets/lmqg/qg_esquad) | - ***Metric (Question & Answer Generation)***: [raw metric file](https://huggingface.co/lmqg/mt5-base-esquad-qg-ae/raw/main/eval/metric.first.answer.paragraph.questions_answers.lmqg_qg_esquad.default.json) | | Score | Type | Dataset | |:--------------------------------|--------:|:--------|:-----------------------------------------------------------------| | BERTScore | 69.46 | default | [lmqg/qg_esquad](https://huggingface.co/datasets/lmqg/qg_esquad) | | Bleu_1 | 10.75 | default | [lmqg/qg_esquad](https://huggingface.co/datasets/lmqg/qg_esquad) | | Bleu_2 | 5.37 | default | [lmqg/qg_esquad](https://huggingface.co/datasets/lmqg/qg_esquad) | | Bleu_3 | 2.92 | default | [lmqg/qg_esquad](https://huggingface.co/datasets/lmqg/qg_esquad) | | Bleu_4 | 1.78 | default | [lmqg/qg_esquad](https://huggingface.co/datasets/lmqg/qg_esquad) | | METEOR | 22.16 | default | [lmqg/qg_esquad](https://huggingface.co/datasets/lmqg/qg_esquad) | | MoverScore | 51.69 | default | [lmqg/qg_esquad](https://huggingface.co/datasets/lmqg/qg_esquad) | | QAAlignedF1Score (BERTScore) | 79.67 | default | [lmqg/qg_esquad](https://huggingface.co/datasets/lmqg/qg_esquad) | | QAAlignedF1Score (MoverScore) | 54.82 | default | [lmqg/qg_esquad](https://huggingface.co/datasets/lmqg/qg_esquad) | | QAAlignedPrecision (BERTScore) | 77.14 | default | [lmqg/qg_esquad](https://huggingface.co/datasets/lmqg/qg_esquad) | | QAAlignedPrecision (MoverScore) | 53.27 | default | [lmqg/qg_esquad](https://huggingface.co/datasets/lmqg/qg_esquad) | | QAAlignedRecall (BERTScore) | 82.44 | default | [lmqg/qg_esquad](https://huggingface.co/datasets/lmqg/qg_esquad) | | QAAlignedRecall (MoverScore) | 56.56 | default | [lmqg/qg_esquad](https://huggingface.co/datasets/lmqg/qg_esquad) | | ROUGE_L | 15.15 | default | [lmqg/qg_esquad](https://huggingface.co/datasets/lmqg/qg_esquad) | - ***Metric (Answer Extraction)***: [raw metric file](https://huggingface.co/lmqg/mt5-base-esquad-qg-ae/raw/main/eval/metric.first.answer.paragraph_sentence.answer.lmqg_qg_esquad.default.json) | | Score | Type | Dataset | |:-----------------|--------:|:--------|:-----------------------------------------------------------------| | AnswerExactMatch | 57.98 | default | [lmqg/qg_esquad](https://huggingface.co/datasets/lmqg/qg_esquad) | | AnswerF1Score | 75.33 | default | [lmqg/qg_esquad](https://huggingface.co/datasets/lmqg/qg_esquad) | | BERTScore | 90.04 | default | [lmqg/qg_esquad](https://huggingface.co/datasets/lmqg/qg_esquad) | | Bleu_1 | 37.35 | default | [lmqg/qg_esquad](https://huggingface.co/datasets/lmqg/qg_esquad) | | Bleu_2 | 32.53 | default | [lmqg/qg_esquad](https://huggingface.co/datasets/lmqg/qg_esquad) | | Bleu_3 | 28.86 | default | [lmqg/qg_esquad](https://huggingface.co/datasets/lmqg/qg_esquad) | | Bleu_4 | 25.75 | default | [lmqg/qg_esquad](https://huggingface.co/datasets/lmqg/qg_esquad) | | METEOR | 43.74 | default | [lmqg/qg_esquad](https://huggingface.co/datasets/lmqg/qg_esquad) | | MoverScore | 80.94 | default | [lmqg/qg_esquad](https://huggingface.co/datasets/lmqg/qg_esquad) | | ROUGE_L | 49.61 | default | [lmqg/qg_esquad](https://huggingface.co/datasets/lmqg/qg_esquad) | ## Training hyperparameters The following hyperparameters were used during fine-tuning: - dataset_path: lmqg/qg_esquad - dataset_name: default - input_types: ['paragraph_answer', 'paragraph_sentence'] - output_types: ['question', 'answer'] - prefix_types: ['qg', 'ae'] - model: google/mt5-base - max_length: 512 - max_length_output: 32 - epoch: 7 - batch: 32 - lr: 0.001 - fp16: False - random_seed: 1 - gradient_accumulation_steps: 2 - label_smoothing: 0.15 The full configuration can be found at [fine-tuning config file](https://huggingface.co/lmqg/mt5-base-esquad-qg-ae/raw/main/trainer_config.json). ## 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", } ```