arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2203.07586cs.CL
收藏

基于自顶向下和自自底向上的推理的长文档摘要

Long Document Summarization with Top-down and Bottom-up Inference

Salesforce Research(赛富时研究院)

Bo Pang, Erik Nijkamp, Wojciech Kryściński, Silvio Savarese, Yingbo Zhou, Caiming Xiong

摘要:

文本摘要旨在压缩长文档并保留关键信息。摘要模型成功的关键是对源文档中单词或标记的潜在表示的忠实推断。最新的模型推断潜在的表示与Transformer编码器,这是纯粹的自自底向上的【形】。此外,基于自我注意的推理模型面临着序列长度的二次复杂性的挑战。我们提出了一个原则性的推理框架,以改善这两个方面的摘要模型。我们的框架假设了一个分层的潜在结构的文档,其中顶层捕获的长距离依赖在一个粗略的时间尺度和底部令牌级别保留的细节。重要的是,这种层次结构使令牌表示能够以自自底向上的【形】和自顶向下的方式更新。在自自底向上的【形】的过程中,使用局部自我注意力来推断令牌表示以利用其效率。然后应用自上而下的校正以允许令牌捕获长范围依赖性。我们证明了所提出的框架在不同的摘要数据集上的有效性,包括叙事,会话,科学文档和新闻。我们的模型实现了(1)与全注意力Transformers相比,在具有更高内存和计算效率的短文档上具有竞争力或更好的性能,以及(2)与最近的高效Transformers相比,在广泛的长文档摘要基准上具有最先进的性能。我们还表明,我们的模型可以总结一整本书,并实现竞争力的性能使用$0.27\%$参数(464 M与175 B)和更少的训练数据,相比,最近的GPT-3为基础的模型。这些结果表明,所提出的框架的普遍适用性和好处。

英文摘要:

Text summarization aims to condense long documents and retain key information. Critical to the success of a summarization model is the faithful inference of latent representations of words or tokens in the source documents. Most recent models infer the latent representations with a transformer encoder, which is purely bottom-up. Also, self-attention-based inference models face the challenge of quadratic complexity with respect to sequence length. We propose a principled inference framework to improve summarization models on these two aspects. Our framework assumes a hierarchical latent structure of a document where the top-level captures the long range dependency at a coarser time scale and the bottom token level preserves the details. Critically, this hierarchical structure enables token representations to be updated in both a bottom-up and top-down manner. In the bottom-up pass, token representations are inferred with local self-attention to leverage its efficiency. Top-down correction is then applied to allow tokens to capture long-range dependency. We demonstrate the effectiveness of the proposed framework on a diverse set of summarization datasets, including narrative, conversational, scientific documents and news. Our model achieves (1) competitive or better performance on short documents with higher memory and compute efficiency, compared to full attention transformers, and (2) state-of-the-art performance on a wide range of long document summarization benchmarks, compared to recent efficient transformers. We also show that our model can summarize an entire book and achieve competitive performance using $0.27\%$ parameters (464M vs. 175B) and much less training data, compared to a recent GPT-3-based model. These results indicate the general applicability and benefits of the proposed framework.

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。

↑