--- license: cc-by-4.0 metrics: - bleu4 - meteor - rouge-l - bertscore - moverscore language: en datasets: - lmqg/qg_squad pipeline_tag: text2text-generation tags: - question generation - answer extraction widget: - text: "generate question: Beyonce further expanded her acting career, starring as blues singer Etta James in the 2008 musical biopic, Cadillac Records." example_title: "Question Generation Example 1" - text: "generate question: Beyonce further expanded her acting career, starring as blues singer Etta James in the 2008 musical biopic, Cadillac Records." example_title: "Question Generation Example 2" - text: "generate question: Beyonce further expanded her acting career, starring as blues singer Etta James in the 2008 musical biopic, Cadillac Records ." example_title: "Question Generation Example 3" - text: "extract answers: Beyonce further expanded her acting career, starring as blues singer Etta James in the 2008 musical biopic, Cadillac Records. Her performance in the film received praise from critics, and she garnered several nominations for her portrayal of James, including a Satellite Award nomination for Best Supporting Actress, and a NAACP Image Award nomination for Outstanding Supporting Actress." example_title: "Answer Extraction Example 1" - text: "extract answers: Beyonce further expanded her acting career, starring as blues singer Etta James in the 2008 musical biopic, Cadillac Records. Her performance in the film received praise from critics, and she garnered several nominations for her portrayal of James, including a Satellite Award nomination for Best Supporting Actress, and a NAACP Image Award nomination for Outstanding Supporting Actress. " example_title: "Answer Extraction Example 2" model-index: - name: lmqg/flan-t5-base-squad-qg-ae results: - task: name: Text2text Generation type: text2text-generation dataset: name: lmqg/qg_squad type: default args: default metrics: - name: BLEU4 (Question Generation) type: bleu4_question_generation value: 26.43 - name: ROUGE-L (Question Generation) type: rouge_l_question_generation value: 53.37 - name: METEOR (Question Generation) type: meteor_question_generation value: 26.99 - name: BERTScore (Question Generation) type: bertscore_question_generation value: 90.61 - name: MoverScore (Question Generation) type: moverscore_question_generation value: 64.75 - name: QAAlignedF1Score-BERTScore (Question & Answer Generation (with Gold Answer)) type: qa_aligned_f1_score_bertscore_question_answer_generation_with_gold_answer value: 93.31 - name: QAAlignedRecall-BERTScore (Question & Answer Generation (with Gold Answer)) type: qa_aligned_recall_bertscore_question_answer_generation_with_gold_answer value: 93.99 - name: QAAlignedPrecision-BERTScore (Question & Answer Generation (with Gold Answer)) type: qa_aligned_precision_bertscore_question_answer_generation_with_gold_answer value: 92.65 - name: QAAlignedF1Score-MoverScore (Question & Answer Generation (with Gold Answer)) type: qa_aligned_f1_score_moverscore_question_answer_generation_with_gold_answer value: 64.59 - name: QAAlignedRecall-MoverScore (Question & Answer Generation (with Gold Answer)) type: qa_aligned_recall_moverscore_question_answer_generation_with_gold_answer value: 65.57 - name: QAAlignedPrecision-MoverScore (Question & Answer Generation (with Gold Answer)) type: qa_aligned_precision_moverscore_question_answer_generation_with_gold_answer value: 63.7 - name: BLEU4 (Answer Extraction) type: bleu4_answer_extraction value: 35.6 - name: ROUGE-L (Answer Extraction) type: rouge_l_answer_extraction value: 68.47 - name: METEOR (Answer Extraction) type: meteor_answer_extraction value: 42.76 - name: BERTScore (Answer Extraction) type: bertscore_answer_extraction value: 91.28 - name: MoverScore (Answer Extraction) type: moverscore_answer_extraction value: 81.31 - name: AnswerF1Score (Answer Extraction) type: answer_f1_score__answer_extraction value: 68.88 - name: AnswerExactMatch (Answer Extraction) type: answer_exact_match_answer_extraction value: 57.39 --- # Model Card of `lmqg/flan-t5-base-squad-qg-ae` This model is fine-tuned version of [google/flan-t5-base](https://huggingface.co/google/flan-t5-base) for question generation and answer extraction jointly on the [lmqg/qg_squad](https://huggingface.co/datasets/lmqg/qg_squad) (dataset_name: default) via [`lmqg`](https://github.com/asahi417/lm-question-generation). ### Overview - **Language model:** [google/flan-t5-base](https://huggingface.co/google/flan-t5-base) - **Language:** en - **Training data:** [lmqg/qg_squad](https://huggingface.co/datasets/lmqg/qg_squad) (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="en", model="lmqg/flan-t5-base-squad-qg-ae") # model prediction question_answer_pairs = model.generate_qa("William Turner was an English painter who specialised in watercolour landscapes") ``` - With `transformers` ```python from transformers import pipeline pipe = pipeline("text2text-generation", "lmqg/flan-t5-base-squad-qg-ae") # answer extraction answer = pipe("generate question: Beyonce further expanded her acting career, starring as blues singer Etta James in the 2008 musical biopic, Cadillac Records.") # question generation question = pipe("extract answers: Beyonce further expanded her acting career, starring as blues singer Etta James in the 2008 musical biopic, Cadillac Records. Her performance in the film received praise from critics, and she garnered several nominations for her portrayal of James, including a Satellite Award nomination for Best Supporting Actress, and a NAACP Image Award nomination for Outstanding Supporting Actress.") ``` ## Evaluation - ***Metric (Question Generation)***: [raw metric file](https://huggingface.co/lmqg/flan-t5-base-squad-qg-ae/raw/main/eval/metric.first.sentence.paragraph_answer.question.lmqg_qg_squad.default.json) | | Score | Type | Dataset | |:-----------|--------:|:--------|:---------------------------------------------------------------| | BERTScore | 90.61 | default | [lmqg/qg_squad](https://huggingface.co/datasets/lmqg/qg_squad) | | Bleu_1 | 58.99 | default | [lmqg/qg_squad](https://huggingface.co/datasets/lmqg/qg_squad) | | Bleu_2 | 42.92 | default | [lmqg/qg_squad](https://huggingface.co/datasets/lmqg/qg_squad) | | Bleu_3 | 33.3 | default | [lmqg/qg_squad](https://huggingface.co/datasets/lmqg/qg_squad) | | Bleu_4 | 26.43 | default | [lmqg/qg_squad](https://huggingface.co/datasets/lmqg/qg_squad) | | METEOR | 26.99 | default | [lmqg/qg_squad](https://huggingface.co/datasets/lmqg/qg_squad) | | MoverScore | 64.75 | default | [lmqg/qg_squad](https://huggingface.co/datasets/lmqg/qg_squad) | | ROUGE_L | 53.37 | default | [lmqg/qg_squad](https://huggingface.co/datasets/lmqg/qg_squad) | - ***Metric (Question & Answer Generation)***: [raw metric file](https://huggingface.co/lmqg/flan-t5-base-squad-qg-ae/raw/main/eval/metric.first.answer.paragraph.questions_answers.lmqg_qg_squad.default.json) | | Score | Type | Dataset | |:--------------------------------|--------:|:--------|:---------------------------------------------------------------| | QAAlignedF1Score (BERTScore) | 93.31 | default | [lmqg/qg_squad](https://huggingface.co/datasets/lmqg/qg_squad) | | QAAlignedF1Score (MoverScore) | 64.59 | default | [lmqg/qg_squad](https://huggingface.co/datasets/lmqg/qg_squad) | | QAAlignedPrecision (BERTScore) | 92.65 | default | [lmqg/qg_squad](https://huggingface.co/datasets/lmqg/qg_squad) | | QAAlignedPrecision (MoverScore) | 63.7 | default | [lmqg/qg_squad](https://huggingface.co/datasets/lmqg/qg_squad) | | QAAlignedRecall (BERTScore) | 93.99 | default | [lmqg/qg_squad](https://huggingface.co/datasets/lmqg/qg_squad) | | QAAlignedRecall (MoverScore) | 65.57 | default | [lmqg/qg_squad](https://huggingface.co/datasets/lmqg/qg_squad) | - ***Metric (Answer Extraction)***: [raw metric file](https://huggingface.co/lmqg/flan-t5-base-squad-qg-ae/raw/main/eval/metric.first.answer.paragraph_sentence.answer.lmqg_qg_squad.default.json) | | Score | Type | Dataset | |:-----------------|--------:|:--------|:---------------------------------------------------------------| | AnswerExactMatch | 57.39 | default | [lmqg/qg_squad](https://huggingface.co/datasets/lmqg/qg_squad) | | AnswerF1Score | 68.88 | default | [lmqg/qg_squad](https://huggingface.co/datasets/lmqg/qg_squad) | | BERTScore | 91.28 | default | [lmqg/qg_squad](https://huggingface.co/datasets/lmqg/qg_squad) | | Bleu_1 | 49.4 | default | [lmqg/qg_squad](https://huggingface.co/datasets/lmqg/qg_squad) | | Bleu_2 | 44.53 | default | [lmqg/qg_squad](https://huggingface.co/datasets/lmqg/qg_squad) | | Bleu_3 | 39.73 | default | [lmqg/qg_squad](https://huggingface.co/datasets/lmqg/qg_squad) | | Bleu_4 | 35.6 | default | [lmqg/qg_squad](https://huggingface.co/datasets/lmqg/qg_squad) | | METEOR | 42.76 | default | [lmqg/qg_squad](https://huggingface.co/datasets/lmqg/qg_squad) | | MoverScore | 81.31 | default | [lmqg/qg_squad](https://huggingface.co/datasets/lmqg/qg_squad) | | ROUGE_L | 68.47 | default | [lmqg/qg_squad](https://huggingface.co/datasets/lmqg/qg_squad) | ## Training hyperparameters The following hyperparameters were used during fine-tuning: - dataset_path: lmqg/qg_squad - dataset_name: default - input_types: ['paragraph_answer', 'paragraph_sentence'] - output_types: ['question', 'answer'] - prefix_types: ['qg', 'ae'] - model: google/flan-t5-base - max_length: 512 - max_length_output: 32 - epoch: 7 - batch: 8 - lr: 0.0001 - fp16: False - random_seed: 1 - gradient_accumulation_steps: 16 - label_smoothing: 0.15 The full configuration can be found at [fine-tuning config file](https://huggingface.co/lmqg/flan-t5-base-squad-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", } ```