Towards Coherent and Cohesive Long-form Text Generation

By Woon Sang Cho, Pengchuan Zhang, Yizhe Zhang, Xiujun Li, Michel Galley, Chris Brockett, Mengdi Wang, Jianfeng Gao. 2018

58 citations recorded by the source corpus.

Generating coherent and cohesive long-form texts is a challenging task. Previous works relied on large amounts of human-generated texts to train neural language models. However, few attempted to explicitly improve neural language models from the perspectives of coherence and cohesion. In this work, we propose a new neural language model that is equipped with two neural discriminators which provide feedback signals at the levels of sentence (cohesion) and paragraph (coherence). Our model is trained using a simple yet efficient variant of policy gradient, called ‘negative-critical sequence training’, which is proposed to eliminate the need of training a separate critic for estimating ‘baseline’. Results demonstrate the effectiveness of our approach, showing improvements over the strong baseline – recurrent attention-based bidirectional MLE-trained neural language model.

Subfields: Natural Language Processing

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Natural Language Processing

Citations in the corpus

Training language models to follow instructions with human feedback (2022, background): “…dialogue (Jaques et al., 2019; Yi et al., 2019; Hancock et al., 2019), translation (Kreutzer et al., 2018; Bahdanau et al., 2016), semantic parsing (Lawrence and Riezler, 2018), story generation (Zhou and Xu, 2020), review generation (Cho et al., 2018), and evidence extraction (Perez et al., 2019).”
Fine-Tuning Language Models from Human Preferences (2019, methodology): “Wu and Hu (2018) and Cho et al. (2019) learned models of coherence from existing text and used them as RL rewards for summarization and long-form generation, respectively.”
Learning to summarize from human feedback (2020, background): “Human feedback has also been used as a reward to train models in other domains such as dialogue [25, 68, 21], translation [32, 1], semantic parsing [34], story generation [72], review generation [7], and evidence extraction [46].”
Recursively Summarizing Books with Human Feedback (2021, background): “…et al., 2018; Bahdanau et al., 2016), semantic parsing (Lawrence and Riezler, 2018), story generation (Zhou and Xu, 2020), review generation (Cho et al., 2018), and evidence extraction (Perez et al., 2019), and agents in simulated environments (Christiano et al., 2017; Ibarz et al., 2018).”
Automatic Document Sketching: Generating Drafts from Analogous Texts (2021, background): “Recent work leverages the success of large pre-trained language models to generate long texts such as stories (Rashkin et al., 2020), reviews (Cho et al., 2019a,b) and fake news (Zellers et al., 2019).”
Story Centaur: Large Language Model Few Shot Learning as a Creative Writing Tool (2021, background): “However, cracks begin to show in coherence as generations grow longer (Cho et al., 2018).”
Moral Stories: Situated Reasoning about Norms, Intents, Actions, and their Consequences (2020, methodology): “The proposed CoE algorithms, on the other hand, are closely related to rescoring methods employed in NLG, including work by (Holtzman et al., 2018; Cho et al., 2019; Gabriel et al., 2019; Hossain et al., 2020; Goldfarb-Tarrant et al., 2020), among others.”
AutoSurvey2: Empowering Researchers with Next Level Automated Literature Surveys (2025, unknown): “Building upon recent advances in automated literature review generation (Bosselut et al., 2018; Cho et al., 2018; Wang et al., 2019), the system integrates large language models (LLMs), semantic vector retrieval, and a modular graph-based execution mechanism to emulate the scholarly writing process…”
AutoSurvey: Large Language Models Can Automatically Write Surveys (2024, unknown): “Compared to traditional methods, which primarily focus on training models to generate linguistically coherent text [39, 40], the emergency of large language models (LLMs) has opened up new possibilities for automated writing, drawing more attention to broader aspects like faithfulness, logical…”
Bloom-Eval: A Hierarchical Evaluation Benchmark for Automatic Survey Generation Based on Bloom's Taxonomy (2026, unknown): “…Wang et al., 2019; Lu et al., 2020; Jin et al., 2020; Xing et al., 2020) and long-form text generation (Fan et al., 2018; Bosselut et al., 2018; Cho et al., 2019; Mao et al., 2022; Kryscinski et al., 2022; Chang et al., 2024), where researchers have explored diverse strategies for information…”