2023/05/24 by Tanay Dixit, Fei Wang, Dixit, Tanay +3 · 2 citations
Computer Science · Engineering · #Advanced Text Analysis Techniques #Artificial Intelligence (cs.AI) #Artificial intelligence #Automatic summarization #Computation and Language (cs.CL) #Computer science #Consistency (knowledge bases) #Engineering #FOS: Computer and information sciences #Image (mathematics) #Information retrieval #Machine Learning (cs.LG) #Machine learning #Metric (unit) #Natural Language Processing Techniques #Natural language processing #Quality (philosophy) #Ranking (information retrieval) #Similarity (geometry) #Topic Modeling
paper · pdf · doi:10.48550/arxiv.2305.14981
published in arXiv (Cornell University) (Cornell University)
openalex publication_date 2023/05/24 · openalex created_date 2023/05/27 · openalex updated_date 2026/07/28
Improving factual consistency of abstractive summarization has been a widely studied topic. However, most of the prior works on training factuality-aware models have ignored the negative effect it has on summary quality. We propose EFACTSUM (i.e., Effective Factual Summarization), a candidate summary generation and ranking technique to improve summary factuality without sacrificing summary quality. We show that using a contrastive learning framework with our refined candidate summaries leads to significant gains on both factuality and similarity-based metrics. Specifically, we propose a ranking strategy in which we effectively combine two metrics, thereby preventing any conflict during training. Models trained using our approach show up to 6 points of absolute improvement over the base model with respect to FactCC on XSUM and 11 points on CNN/DM, without negatively affecting either similarity-based metrics or absractiveness.