Can Large Language Models Follow Concept Annotation Guidelines? A Case Study on Scientific and Financial Domains

Marcio Fonseca, Shay Cohen

Abstract
Although large language models (LLMs) exhibit remarkable capacity to leverage in-context demonstrations, it is still unclear to what extent they can learn new facts or concept definitions via prompts. To address this question, we examine the capacity of instruction-tuned LLMs to follow in-context concept annotation guidelines for zero-shot sentence labeling tasks. We design guidelines that present different types of factual and counterfactual concept definitions, which are used as prompts for zero-shot sentence classification tasks. Our results show that although concept definitions consistently help in task performance, only the larger models (with 70B parameters or more) have limited ability to work under counterfactual contexts. Importantly, only proprietary models such as GPT-3.5 can recognize nonsensical guidelines, which we hypothesize is due to more sophisticated alignment methods. Finally, we find that Falcon-180B-chat is outperformed by Llama-2-70B-chat is most cases, which indicates that increasing model scale does not guarantee better adherence to guidelines. Altogether, our simple evaluation method reveals significant gaps in concept understanding between the most capable open-source language models and the leading proprietary APIs.
Anthology ID:
2024.findings-acl.478
Volume:
Findings of the Association for Computational Linguistics ACL 2024
Month:
August
Year:
2024
Address:
Bangkok, Thailand and virtual meeting
Editors:
Lun-Wei Ku, Andre Martins, Vivek Srikumar
Venue:
Findings
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
8027–8042
Language:
URL:
https://aclanthology.org/2024.findings-acl.478
DOI:
10.18653/v1/2024.findings-acl.478
Bibkey:
Cite (ACL):
Marcio Fonseca and Shay Cohen. 2024. Can Large Language Models Follow Concept Annotation Guidelines? A Case Study on Scientific and Financial Domains. In Findings of the Association for Computational Linguistics ACL 2024, pages 8027–8042, Bangkok, Thailand and virtual meeting. Association for Computational Linguistics.
Cite (Informal):
Can Large Language Models Follow Concept Annotation Guidelines? A Case Study on Scientific and Financial Domains (Fonseca & Cohen, Findings 2024)
Copy Citation:
PDF:
https://preview.aclanthology.org/ingest-2024-clasp/2024.findings-acl.478.pdf
PDF Search