Lune

ACL2022Top-tier venue

A Multi-Document Coverage Reward for RELAXed Multi-Document Summarization

Jacob Parnell, Inigo Jauregi Unanue, Massimo Piccardi

Abstract

Multi-document summarization (MDS) has made significant progress in recent years, in part facilitated by the availability of new, dedicated datasets and capacious language models. However, a standing limitation of these models is that they are trained against limited references and with plain maximum-likelihood objectives. As for many other generative tasks, reinforcement learning (RL) offers the potential to improve the training of MDS models; yet, it requires a carefully-designed reward that can ensure appropriate leverage of both the reference summaries and the input documents. For this reason, in this paper we propose fine-tuning an MDS baseline with a reward that balances a reference-based metric such as ROUGE with coverage of the input documents. To implement the approach, we utilize RELAX (Grathwohl et al., 2018) , a contemporary gradient estimator which is both low-variance and unbiased, and we fine-tune the baseline in a few-shot style for both stability and computational efficiency. Experimental results over the Multi-News and WCEP MDS datasets show significant improvements of up to +0.95 pp average ROUGE score and +3.17 pp METEOR score over the baseline, and competitive results with the literature. In addition, they show that the coverage of the input documents is increased, and evenly across all documents.

Ask Lune

Know this paper in minutes, not an afternoon.

Lune can read every page of it and the top-tier papers that cite it, and gives you its claims, its evidence and its limits, each one cited to the paper it comes from. No skimming, no AI slop.

2 top-tier papers have cited it since. Know what they found before your reviewers do.

Questions to start from

Lune can draw on

the full paper · the top-tier papers that cite it · related work

Get the answer

Free to start. No credit card required.

Lineage

  • Big Bird: Transformers for Longer Sequences cites PEGASUS: Pre-training with Extracted Gap-sentences for Abstractive Summarization
  • Multi-Dimensional Optimization for Text Summarization via Reinforcement Learning cites BART: Denoising Sequence-to-Sequence Pre-training for Natural Language Generation, Translation, and Comprehension
  • Multi-Dimensional Optimization for Text Summarization via Reinforcement Learning cites BERTScore: Evaluating Text Generation with BERT
  • Multi-Dimensional Optimization for Text Summarization via Reinforcement Learning cites PEGASUS: Pre-training with Extracted Gap-sentences for Abstractive Summarization
  • Autoregressive Multi-trait Essay Scoring via Reinforcement Learning with Scoring-aware Multiple Rewards cites Multi-Dimensional Optimization for Text Summarization via Reinforcement Learning
Builds onCited byLeft to right by year; the more cited, the higher.

Cited by top-tier papers2

  1. Autoregressive Multi-trait Essay Scoring via Reinforcement Learning with Scoring-aware Multiple RewardsHeejin Do, Sangwon Ryu, Gary Geunbae LeeEMNLP 2024 · 5 citations
  2. Multi-Dimensional Optimization for Text Summarization via Reinforcement LearningSangwon Ryu, Heejin Do, Yunsu Kim, Gary Lee et al.ACL 2024

Builds on6

  1. BERTScore: Evaluating Text Generation with BERTTianyi Zhang, Varsha Kishore, Felix Wu, Kilian Q. Weinberger et al.ICLR 2020 · 8,443 citations
  2. Big Bird: Transformers for Longer SequencesManzil Zaheer, Guru Guruganesh, Kumar Avinava Dubey, Joshua Ainslie et al.NeurIPS 2020 · 3,159 citations
  3. PEGASUS: Pre-training with Extracted Gap-sentences for Abstractive SummarizationJingqing Zhang, Yao Zhao, Mohammad Saleh, Peter J. LiuICML 2020 · 2,453 citations
  4. BART: Denoising Sequence-to-Sequence Pre-training for Natural Language Generation, Translation, and ComprehensionMike Lewis, Yinhan Liu, Naman Goyal, Marjan Ghazvininejad et al.ACL 2020 · 1,224 citations
  5. Leveraging Graph to Improve Abstractive Multi-Document SummarizationWei Li, Xinyan Xiao, Jiachen Liu, Hua Wu et al.ACL 2020 · 118 citations

Related papers

  1. Topic-Guided Reinforcement Learning with LLMs for Enhancing Multi-Document SummarizationChuyuan Li, Austin Xu, Shafiq Joty, Giuseppe CareniniEMNLP Findings 2025
  2. Multi-document Summarization with Maximal Marginal Relevance-guided Reinforcement LearningYuning Mao, Yanru Qu, Yiqing Xie, Xiang Ren et al.EMNLP 2020 · 43 citations
  3. Learning to summarize with human feedbackNisan Stiennon, Long Ouyang, Jeffrey Wu, Daniel M. Ziegler et al.NeurIPS 2020 · 124 citations
  4. Topic-Guided Abstractive Multi-Document SummarizationPeng Cui, Le HuEMNLP Findings 2021 · 38 citations
  5. Training Diffusion Models Towards Diverse Image Generation with Reinforcement LearningZichen Miao, Jiang Wang, Ze Wang, Zhengyuan Yang et al.CVPR 2024 · 12 citations
Dusk over the sea between two cliffs drawn in fine vertical lines