@inproceedings{jain-etal-2024-natural,
    title = "Natural Answer Generation: From Factoid Answer to Full-length Answer using Grammar Correction",
    author = "Jain, Manas  and
      Saha, Sriparna  and
      Bhattacharyya, Pushpak  and
      Chinnadurai, Gladvin  and
      Vatsa, Manish",
    editor = "Lalitha Devi, Sobha  and
      Arora, Karunesh",
    booktitle = "Proceedings of the 21st International Conference on Natural Language Processing (ICON)",
    month = dec,
    year = "2024",
    address = "AU-KBC Research Centre, Chennai, India",
    publisher = "NLP Association of India (NLPAI)",
    url = "https://preview.aclanthology.org/fix-sig-urls/2024.icon-1.43/",
    pages = "376--385",
    abstract = "Question Answering systems these days typically use template-based language generation. Though adequate for a domain-specific task, these systems are too restrictive and predefined for domain-independent systems. This paper proposes a system that outputs a full-length answer given a question and the extracted factoid answer (short spans such as named entities) as the input. Our system uses constituency and dependency parse trees of questions. A transformer-based Grammar Error Correction model GECToR is used as a post-processing step for better fluency. We compare our system with (i) a Modified Pointer Generator (SOTA) and (ii) Fine-tuned DialoGPT for factoid questions. We also tested our approach on existential (yes-no) questions with better results. Our model generates more accurate and fluent answers than the state-of-the-art (SOTA) approaches. The evaluation is done on NewsQA and SqUAD datasets with an increment of 0.4 and 0.9 percentage points in ROUGE-1 score respectively. Also, the inference time is reduced by 85{\%} compared to the SOTA. The improved datasets used for our evaluation will be released as part of the research contribution."
}