Abstract
Retrieval-augmented methods are successful in the standard scenario where the retrieval space is sufficient; whereas in the few-shot scenario with limited retrieval space, this paper shows it is non-trivial to put them into practice. First, it is impossible to retrieve semantically similar examples by using an off-the-shelf metric and it is crucial to learn a task-specific retrieval metric; Second, our preliminary experiments demonstrate that it is difficult to optimize a plausible metric by minimizing the standard cross-entropy loss. The in-depth analyses quantitatively show minimizing cross-entropy loss suffers from the weak supervision signals and the severe gradient vanishing issue during the optimization. To address these issues, we introduce two novel training objectives, namely EM-L and R-L, which provide more task-specific guidance to the retrieval metric by the EM algorithm and a ranking-based loss, respectively. Extensive experiments on 10 datasets prove the superiority of the proposed retrieval augmented methods on the performance.- Anthology ID:
- 2023.findings-emnlp.447
- 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:
- 6721–6735
- Language:
- URL:
- https://preview.aclanthology.org/icon-24-ingestion/2023.findings-emnlp.447/
- DOI:
- 10.18653/v1/2023.findings-emnlp.447
- Cite (ACL):
- Guoxin Yu, Lemao Liu, Haiyun Jiang, Shuming Shi, and Xiang Ao. 2023. Retrieval-Augmented Few-shot Text Classification. In Findings of the Association for Computational Linguistics: EMNLP 2023, pages 6721–6735, Singapore. Association for Computational Linguistics.
- Cite (Informal):
- Retrieval-Augmented Few-shot Text Classification (Yu et al., Findings 2023)
- PDF:
- https://preview.aclanthology.org/icon-24-ingestion/2023.findings-emnlp.447.pdf