Exploring and Mitigating Shortcut Learning for Generative Large Language Models
Zechen Sun, Yisheng Xiao, Juntao Li, Yixin Ji, Wenliang Chen, Min Zhang
Abstract
Recent generative large language models (LLMs) have exhibited incredible instruction-following capabilities while keeping strong task completion ability, even without task-specific fine-tuning. Some works attribute this to the bonus of the new scaling law, in which the continuous improvement of model capacity yields emergent capabilities, e.g., reasoning and universal generalization. However, we point out that recent LLMs still show shortcut learning behavior, where the models tend to exploit spurious correlations between non-robust features and labels for prediction, which might lead to overestimating model capabilities. LLMs memorize more complex spurious correlations (i.e., task ↔ feature ↔ label) compared with that learned from previous pre-training and task-specific fine-tuning paradigm (i.e., feature ↔ label). Based on our findings, we propose FSLI, a framework for encouraging LLMs to Forget Spurious correlations and Learn from In-context information. Experiments on three tasks show that FSFI can effectively mitigate shortcut learning. Besides, we argue not to overestimate the capabilities of LLMs and conduct evaluations in more challenging and complete test scenarios.- Anthology ID:
- 2024.lrec-main.602
- Volume:
- Proceedings of the 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation (LREC-COLING 2024)
- Month:
- May
- Year:
- 2024
- Address:
- Torino, Italia
- Editors:
- Nicoletta Calzolari, Min-Yen Kan, Veronique Hoste, Alessandro Lenci, Sakriani Sakti, Nianwen Xue
- Venues:
- LREC | COLING
- SIG:
- Publisher:
- ELRA and ICCL
- Note:
- Pages:
- 6883–6893
- Language:
- URL:
- https://aclanthology.org/2024.lrec-main.602
- DOI:
- Cite (ACL):
- Zechen Sun, Yisheng Xiao, Juntao Li, Yixin Ji, Wenliang Chen, and Min Zhang. 2024. Exploring and Mitigating Shortcut Learning for Generative Large Language Models. In Proceedings of the 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation (LREC-COLING 2024), pages 6883–6893, Torino, Italia. ELRA and ICCL.
- Cite (Informal):
- Exploring and Mitigating Shortcut Learning for Generative Large Language Models (Sun et al., LREC-COLING 2024)
- PDF:
- https://preview.aclanthology.org/nschneid-patch-4/2024.lrec-main.602.pdf