CUED_在TREC 2020播客总结跟踪上的讲话
CUED_speech at TREC 2020 Podcast Summarisation Track
摘要：
在本文中，我们描述了我们在TREC 2020中应对播客摘要挑战的方法。给定一个播客片段及其转录，目标是生成一个摘要，捕获内容中最重要的信息。我们的方法包括两个步骤：（1）使用层次模型的注意力过滤转录中冗余或信息量较少的句子;（2）应用最先进的文本摘要系统（BART），使用序列级奖励函数对播客数据进行微调。此外，我们为提交运行执行三个和九个模型的合奏。我们还对播客数据的BART模型进行了微调，作为我们的基线。NIST的人类评估显示，我们最好的提交在EGFB量表中达到1.777，而创作者提供的描述得分为1.291。我们的系统在TREC 2020播客赛道中赢得了Spotify播客摘要挑战赛的人工和自动评估。
英文摘要：
In this paper, we describe our approach for the Podcast Summarisation challenge in TREC 2020. Given a podcast episode with its transcription, the goal is to generate a summary that captures the most important information in the content. Our approach consists of two steps: (1) Filtering redundant or less informative sentences in the transcription using the attention of a hierarchical model; (2) Applying a state-of-the-art text summarisation system (BART) fine-tuned on the Podcast data using a sequence-level reward function. Furthermore, we perform ensembles of three and nine models for our submission runs. We also fine-tune the BART model on the Podcast data as our baseline. The human evaluation by NIST shows that our best submission achieves 1.777 in the EGFB scale, while the score of creator-provided description is 1.291. Our system won the Spotify Podcast Summarisation Challenge in the TREC2020 Podcast Track in both human and automatic evaluation.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
