The student becomes the master: Outperforming GPT3 on Scientific Factual Error Correction

Dhananjay Ashok, Atharva Kulkarni, Hai Pham, Barnabas Poczos


Abstract
Due to the prohibitively high cost of creating error correction datasets, most Factual Claim Correction methods rely on a powerful verification model to guide the correction process. This leads to a significant drop in performance in domains like Scientific Claim Correction, where good verification models do not always exist. In this work we introduce SciFix, a claim correction system that does not require a verifier but is able to outperform existing methods by a considerable margin — achieving correction accuracy of 84% on the SciFact dataset, 77% on SciFact-Open and 72.75% on the CovidFact dataset, compared to next best accuracies of 7.6%, 5% and 15% on the same datasets respectively. Our method leverages the power of prompting with LLMs during training to create a richly annotated dataset that can be used for fully supervised training and regularization. We additionally use a claim-aware decoding procedure to improve the quality of corrected claims. Our method outperforms the very LLM that was used to generate the annotated dataset — with FewShot Prompting on GPT3.5 achieving 58%, 61% and 64% on the respective datasets, a consistently lower correction accuracy, despite using nearly 800 times as many parameters as our model.
Anthology ID:
2023.findings-emnlp.451
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:
6762–6778
Language:
URL:
https://aclanthology.org/2023.findings-emnlp.451
DOI:
10.18653/v1/2023.findings-emnlp.451
Bibkey:
Cite (ACL):
Dhananjay Ashok, Atharva Kulkarni, Hai Pham, and Barnabas Poczos. 2023. The student becomes the master: Outperforming GPT3 on Scientific Factual Error Correction. In Findings of the Association for Computational Linguistics: EMNLP 2023, pages 6762–6778, Singapore. Association for Computational Linguistics.
Cite (Informal):
The student becomes the master: Outperforming GPT3 on Scientific Factual Error Correction (Ashok et al., Findings 2023)
Copy Citation:
PDF:
https://aclanthology.org/2023.findings-emnlp.451.pdf