Deconstructing In-Context Learning: Understanding Prompts via Corruption
Namrata Shivagunde, Vladislav Lialin, Sherin Muckatira, Anna Rumshisky
Abstract
The ability of large language models (LLMs) to “learn in context” based on the provided prompt has led to an explosive growth in their use, culminating in the proliferation of AI assistants such as ChatGPT, Claude, and Bard. These AI assistants are known to be robust to minor prompt modifications, mostly due to alignment techniques that use human feedback. In contrast, the underlying pre-trained LLMs they use as a backbone are known to be brittle in this respect. Building high-quality backbone models remains a core challenge, and a common approach to assessing their quality is to conduct few-shot evaluation. Such evaluation is notorious for being highly sensitive to minor prompt modifications, as well as the choice of specific in-context examples. Prior work has examined how modifying different elements of the prompt can affect model performance. However, these earlier studies tended to concentrate on a limited number of specific prompt attributes and often produced contradictory results. Additionally, previous research either focused on models with fewer than 15 billion parameters or exclusively examined black-box models like GPT-3 or PaLM, making replication challenging. In the present study, we decompose the entire prompt into four components: task description, demonstration inputs, labels, and inline instructions provided for each demonstration. We investigate the effects of structural and semantic corruptions of these elements on model performance. We study models ranging from 1.5B to 70B in size, using ten datasets covering classification and generation tasks. We find that repeating text within the prompt boosts model performance, and bigger models (≥30B) are more sensitive to the semantics of the prompt. Finally, we observe that adding task and inline instructions to the demonstrations enhances model performance even when the instructions are semantically corrupted. The code is available at this URL.- Anthology ID:
- 2024.lrec-main.404
- 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:
- 4509–4529
- Language:
- URL:
- https://aclanthology.org/2024.lrec-main.404
- DOI:
- Cite (ACL):
- Namrata Shivagunde, Vladislav Lialin, Sherin Muckatira, and Anna Rumshisky. 2024. Deconstructing In-Context Learning: Understanding Prompts via Corruption. In Proceedings of the 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation (LREC-COLING 2024), pages 4509–4529, Torino, Italia. ELRA and ICCL.
- Cite (Informal):
- Deconstructing In-Context Learning: Understanding Prompts via Corruption (Shivagunde et al., LREC-COLING 2024)
- PDF:
- https://preview.aclanthology.org/landing_page/2024.lrec-main.404.pdf