Abstract
We present PeerSum, a novel dataset for generating meta-reviews of scientific papers. The meta-reviews can be interpreted as abstractive summaries of reviews, multi-turn discussions and the paper abstract. These source documents have a rich inter-document relationship with an explicit hierarchical conversational structure, cross-references and (occasionally) conflicting information. To introduce the structural inductive bias into pre-trained language models, we introduce RAMMER (Relationship-aware Multi-task Meta-review Generator), a model that uses sparse attention based on the conversational structure and a multi-task training objective that predicts metadata features (e.g., review ratings). Our experimental results show that RAMMER outperforms other strong baseline models in terms of a suite of automatic evaluation metrics. Further analyses, however, reveal that RAMMER and other models struggle to handle conflicts in source documents, suggesting meta-review generation is a challenging task and a promising avenue for further research.- Anthology ID:
- 2023.findings-emnlp.472
- Volume:
- Findings of the Association for Computational Linguistics: EMNLP 2023
- Month:
- December
- Year:
- 2023
- Address:
- Singapore
- Editors:
- Houda Bouamor, Juan Pino, Kalika Bali
- Venue:
- Findings
- SIG:
- Publisher:
- Association for Computational Linguistics
- Note:
- Pages:
- 7089–7112
- Language:
- URL:
- https://aclanthology.org/2023.findings-emnlp.472
- DOI:
- 10.18653/v1/2023.findings-emnlp.472
- Cite (ACL):
- Miao Li, Eduard Hovy, and Jey Lau. 2023. Summarizing Multiple Documents with Conversational Structure for Meta-Review Generation. In Findings of the Association for Computational Linguistics: EMNLP 2023, pages 7089–7112, Singapore. Association for Computational Linguistics.
- Cite (Informal):
- Summarizing Multiple Documents with Conversational Structure for Meta-Review Generation (Li et al., Findings 2023)
- PDF:
- https://preview.aclanthology.org/emnlp-22-attachments/2023.findings-emnlp.472.pdf