Temporal Blind Spots in Large Language Models

Temporal Blind Spots in Large Language Models

Wallat, Jonas (L3S)
Jatowt, Adam (University of Innsbruck)
Anand, A. (TU Delft Web Information Systems)

2024

Large language models (LLMs) have recently gained significant attention due to their unparalleled zero-shot performance on various natural language processing tasks. However, the pre-Training data utilized in LLMs is often confined to a specific corpus, resulting in inherent freshness and temporal scope limitations. Consequently, this raises concerns regarding the effectiveness of LLMs for tasks involving temporal intents. In this study, we aim to investigate the underlying limitations of general-purpose LLMs when deployed for tasks that require a temporal understanding. We pay particular attention to handling factual temporal knowledge through three popular temporal QA datasets. Specifically, we observe low performance on detailed questions about the past and, surprisingly, for rather new information. In manual and automatic testing, we find multiple temporal errors and characterize the conditions under which QA performance deteriorates. Our analysis contributes to understanding LLM limitations and offers valuable insights into developing future models that can better cater to the demands of temporally-oriented tasks. The code is available https://github.com/jwallat/temporalblindspots.

large language models
question answering
temporal information retrieval
temporal query intents

http://resolver.tudelft.nl/uuid:5a51171d-ec31-4c1b-bdda-c09254704b4d

https://doi.org/10.1145/3616855.3635818

Association for Computing Machinery (ACM), New York

2024-10-14

979-8-4007-0371-3

WSDM 2024 - Proceedings of the 17th ACM International Conference on Web Search and Data Mining

17th ACM International Conference on Web Search and Data Mining, WSDM 2024, 2024-03-04 → 2024-03-08, Merida, Mexico

Green Open Access added to TU Delft Institutional Repository ‘You share, we take care!’ – Taverne project https://www.openaccess.nl/en/you-share-we-take-care Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public.

Institutional Repository

conference paper

© 2024 Jonas Wallat, Adam Jatowt, A. Anand