A Study into Investigating Temporal Robustness of LLMs

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Wallat, Jonas, Abdallah, Abdelrahman, Jatowt, Adam, Anand, Avishek
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866908276912816128
author Wallat, Jonas
Abdallah, Abdelrahman
Jatowt, Adam
Anand, Avishek
author_facet Wallat, Jonas
Abdallah, Abdelrahman
Jatowt, Adam
Anand, Avishek
contents Large Language Models (LLMs) encapsulate a surprising amount of factual world knowledge. However, their performance on temporal questions and historical knowledge is limited because they often cannot understand temporal scope and orientation or neglect the temporal aspect altogether. In this study, we aim to measure precisely how robust LLMs are for question answering based on their ability to process temporal information and perform tasks requiring temporal reasoning and temporal factual knowledge. Specifically, we design eight time-sensitive robustness tests for factual information to check the sensitivity of six popular LLMs in the zero-shot setting. Overall, we find LLMs lacking temporal robustness, especially to temporal reformulations and the use of different granularities of temporal references. We show how a selection of these eight tests can be used automatically to judge a model's temporal robustness for user questions on the fly. Finally, we apply the findings of this study to improve the temporal QA performance by up to 55 percent.
format Preprint
id arxiv_https___arxiv_org_abs_2503_17073
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle A Study into Investigating Temporal Robustness of LLMs
Wallat, Jonas
Abdallah, Abdelrahman
Jatowt, Adam
Anand, Avishek
Computation and Language
Information Retrieval
68T50
I.2.7
Large Language Models (LLMs) encapsulate a surprising amount of factual world knowledge. However, their performance on temporal questions and historical knowledge is limited because they often cannot understand temporal scope and orientation or neglect the temporal aspect altogether. In this study, we aim to measure precisely how robust LLMs are for question answering based on their ability to process temporal information and perform tasks requiring temporal reasoning and temporal factual knowledge. Specifically, we design eight time-sensitive robustness tests for factual information to check the sensitivity of six popular LLMs in the zero-shot setting. Overall, we find LLMs lacking temporal robustness, especially to temporal reformulations and the use of different granularities of temporal references. We show how a selection of these eight tests can be used automatically to judge a model's temporal robustness for user questions on the fly. Finally, we apply the findings of this study to improve the temporal QA performance by up to 55 percent.
title A Study into Investigating Temporal Robustness of LLMs
topic Computation and Language
Information Retrieval
68T50
I.2.7
url https://arxiv.org/abs/2503.17073