Abstract
Evaluating automatically-generated text summaries is a challenging task. While there have been many interesting approaches, they still fall short of human evaluations. We present RISE, a new approach for evaluating summaries by leveraging techniques from information retrieval. RISE is first trained as a retrieval task using a dual-encoder retrieval setup, and can then be subsequently utilized for evaluating a generated summary given an input document, without gold reference summaries. RISE is especially well suited when working on new datasets where one may not have reference summaries available for evaluation. We conduct comprehensive experiments on the SummEval benchmark (Fabbri et al., 2021) and a long document summarization benchmark. The results show that RISE consistently achieves higher correlation with human evaluations compared to many past approaches to summarization evaluation. Furthermore, RISE also demonstrates data-efficiency and generalizability across languages.- Anthology ID:
- 2023.findings-acl.865
- Volume:
- Findings of the Association for Computational Linguistics: ACL 2023
- Month:
- July
- Year:
- 2023
- Address:
- Toronto, Canada
- Editors:
- Anna Rogers, Jordan Boyd-Graber, Naoaki Okazaki
- Venue:
- Findings
- SIG:
- Publisher:
- Association for Computational Linguistics
- Note:
- Pages:
- 13697–13709
- Language:
- URL:
- https://aclanthology.org/2023.findings-acl.865
- DOI:
- 10.18653/v1/2023.findings-acl.865
- Cite (ACL):
- David Uthus and Jianmo Ni. 2023. RISE: Leveraging Retrieval Techniques for Summarization Evaluation. In Findings of the Association for Computational Linguistics: ACL 2023, pages 13697–13709, Toronto, Canada. Association for Computational Linguistics.
- Cite (Informal):
- RISE: Leveraging Retrieval Techniques for Summarization Evaluation (Uthus & Ni, Findings 2023)
- PDF:
- https://preview.aclanthology.org/nschneid-patch-2/2023.findings-acl.865.pdf