paper-with-me

홈 › Papers

Generating Self-Contained and Summary-Centric Question Answer Pairs via Differentiable Reward Imitation Learning

2021-09-10 · EMNLP 2021 11 · Li Zhou, Kevin Small, Yong Zhang, Sandeep Atluri

Motivated by suggested question generation in conversational news recommendation systems, we propose a model for generating question-answer pairs (QA pairs) with self-contained, summary-centric questions and length-constrained, article-summarizing answers. We begin by collecting a new dataset of news articles with questions as titles and pairing them with summaries of varying length. This dataset is used to learn a QA pair generation model producing summaries as answers that balance brevity with sufficiency jointly with their corresponding questions. We then reinforce the QA pair generation process with a differentiable reward function to mitigate exposure bias, a common problem in natural language generation. Both automatic metrics and human evaluation demonstrate these QA pairs successfully capture the central gists of the articles and achieve high answer accuracy.

📄 PDF Abstract BibTeX arXiv:2109.04689

Code (1)

amazon-research/sc2qa-dril 공식 구현 pytorch

Tasks

ArticlesImitation LearningNews RecommendationQuestion GenerationQuestion-GenerationRecommendation SystemsText Generation

Similar Papers 제목 키워드 기반

Educational Question Generation of Children Storybooks via Question Type Distribution Learning and Event-Centric Summarization

2022-03-27 · ACL 2022 5 · Zhenjie Zhao, Yufang Hou, Dakuo Wang, Mo Yu 외

Generating educational questions of fairytales or storybooks is vital for improving children's literacy ability. However, it is challenging to generate questions that capture the interesting aspects of a fairytale story …

Question AnsweringQuestion GenerationQuestion-Generation

Memory-Augmented Multimodal LLMs for Surgical VQA via Self-Contained Inquiry

2024-11-17 · Wenjun Hou, Yi Cheng, Kaishuai Xu, Yan Hu 외

Comprehensively understanding surgical scenes in Surgical Visual Question Answering (Surgical VQA) requires reasoning over multiple objects. Previous approaches address this task using cross-modal fusion strategies to en…

Question AnsweringScene UnderstandingVisual Question AnsweringVisual Question Answering (VQA)

EgoCITE: Context-Augmented Indexing and Time-Aware Retrieval for Long-Horizon Egocentric Memory

2026-08-12 · Le Zhang, Hao Chen, Vlad Roznyatovskiy, Jianzhong Zhang 외 arxiv

Long-horizon egocentric memory transforms continuous first-person video and audio into a searchable record of past experiences. We demonstrate two bottlenecks in existing systems: indices built from context-poor captions…

Generating Multiple-Length Summaries via Reinforcement Learning for Unsupervised Sentence Summarization

2022-12-21 · Dongmin Hyun, Xiting Wang, Chanyoung Park, Xing Xie 외

Sentence summarization shortens given texts while maintaining core contents of the texts. Unsupervised approaches have been studied to summarize texts without human-written summaries. However, recent unsupervised models …

Abstractive Text Summarizationreinforcement-learningReinforcement Learning (RL)Sentence+2

Question-Interlocutor Scope Realized Graph Modeling over Key Utterances for Dialogue Reading Comprehension

2022-10-26 · Jiangnan Li, Mo Yu, Fandong Meng, Zheng Lin 외

In this work, we focus on dialogue reading comprehension (DRC), a task extracting answer spans for questions from dialogues. Dialogue context modeling in DRC is tricky due to complex speaker information and noisy dialogu…

Reading Comprehension