GSum: A General Framework for Guided Neural Abstractive Summarization
By Zi-Yi Dou, Pengfei Liu, Hiroaki Hayashi, Zhengbao Jiang, Graham Neubig. 2020
294 citations recorded by the source corpus.
Neural abstractive summarization models are flexible and can produce coherent summaries, but they are sometimes unfaithful and can be difficult to control. While previous studies attempt to provide different types of guidance to control the output and increase faithfulness, it is not clear how these strategies compare and contrast to each other. In this paper, we propose a general and extensible guided summarization framework (GSum) that can effectively take different kinds of external guidance as input, and we perform experiments across several different varieties. Experiments demonstrate that this model is effective, achieving state-of-the-art performance according to ROUGE on 4 popular summarization datasets when using highlighted sentences as guidance. In addition, we show that our guided model can generate more faithful summaries and demonstrate how different types of guidance generate qualitatively different summaries, lending a degree of controllability to the learned models.
Subfields: Artificial Intelligence, Summarization
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Artificial Intelligence
- Graham Neubig
- Mohit Bansal
- Caiming Xiong
- Pengfei Liu
- Jianfeng Gao
- Percy Liang
- Luke S. Zettlemoyer
- Richard Socher
In Summarization
- Yixin Liu
- Pengfei Liu
- Graham Neubig
- Peng Liu
- Weizhe Yuan
- Dragomir R. Radev
- Bryan McCann
- Shashi Narayan
Citations in the corpus
- Enabling Multimodal Generation on CLIP via Vision-Language Knowledge Distillation (2022, methodology): “Additionally, as presented in Table 5, we also run VLKD on the abstractive summarization task to evaluate its NLG performance, since BART-based methods excel on the summarization (Lewis et al., 2020; Dou et al., 2021; Yu et al., 2021b).”
- BARTScore: Evaluating Generated Text as Text Generation (2021, methodology): “In this work, we operationalize our idea by using BART [32] as our backbone due to its superior performance in text generation [12,42,72].”
- CTRLsum: Towards Generic Controllable Text Summarization (2020, methodology): “Such a (unfair) parameter advantage setting is also adopted in recent two-stage summarization work (Dou et al., 2021).”
- Planning with Learned Entity Prompts for Abstractive Summarization (2021, methodology): “We compare the decoding strategy in FROST to encoder-guided control summarization systems (He et al., 2020; Dou et al., 2021).”
- BRIO: Bringing Order to Abstractive Summarization (2022, methodology): “GSum (Dou et al., 2021) is built on BART, and improves performance by using additional guidance from an extractive summarizer.”
- A Prompting-based Approach for Adversarial Example Generation and Robustness Enhancement (2022, background): “, 2021], natural language inference [Schick and Schütze, 2021], summarization [Li and Liang, 2021; Dou et al., 2021], dialogue systems [Lee et al.”
- RefSum: Refactoring Neural Summarization (2021, methodology): “GSum (Dou et al., 2020) enhances the performance of BART using additional guidance information, which achieves the current state-of-the-art performance on the CNNDM dataset.”
- SimCLS: A Simple Framework for Contrastive Learning of Abstractive Summarization (2021, methodology): “We use BART, Pegasus, GSum (Dou et al., 2021) and ProphetNet (Qi et al.”
- Control Prefixes for Parameter-Efficient Text Generation (2021, background): “GSum (Dou et al., 2020) is an example of work that has explored using learned guidance prediction models at test time.”
- Pre-Trained Language Models for Text Generation: A Survey (2022, background): “GSum [38] proposed a general framework taking different kinds of guidance signals into the generation model, including keywords, triples, highlighted sentences and retrieved summaries.”
- Control Prefixes for Text Generation (2021, background): “A sample size of 300 is typically much larger than that where authors construct their own evaluation (Narayan et al. (2018) use 50 and Dou et al. (2020) use 100).”
- Revisiting the Gold Standard: Grounding Summarization Evaluation with Robust Human Evaluation (2022, background): “For example, the length of GSum (Dou et al., 2021) is around 40% longer than GLOBAL (Ma et al., 2021).”
- How well do you know your summarization datasets? (2021, unknown): “The past few years have witnessed major break-throughs and improvements in automatic summarization (See et al., 2017; Celikyilmaz et al., 2018; Jadhav and Rajan, 2018; Liu and Lapata, 2019; Liu, 2019; Dou et al., 2020; Yuan et al., 2021; Liu et al., 2021).”
- Can We Automate Scientific Reviewing? (2021, methodology): “Instead of regarding text generation as a holistic process, we decompose it into two steps, using an extract-then-generate paradigm (Chen and Bansal, 2018; Gehrmann et al., 2018; Subramanian et al., 2019; Dou et al., 2020).”