English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Large language models (LLMs) have shown nearly saturated performance on manynatural language processing (NLP) tasks. As a result, it is natural for peopleto believe that LLMs have also mastered abilities such as time understandingand reasoning. However, research on the temporal sensitivity of LLMs has beeninsufficiently emphasized. To fill this gap, this paper constructs MultipleSensitive Factors Time QA (MenatQA), which encompasses three temporal factors(scope factor, order factor, counterfactual factor) with total 2,853 samplesfor evaluating the time comprehension and reasoning abilities of LLMs. Thispaper tests current mainstream LLMs with different parameter sizes, rangingfrom billions to hundreds of billions. The results show most LLMs fall behindsmaller temporal reasoning models with different degree on these factors. Inspecific, LLMs show a significant vulnerability to temporal biases and dependheavily on the temporal information provided in questions. Furthermore, thispaper undertakes a preliminary investigation into potential improvementstrategies by devising specific prompts and leveraging external tools. Theseapproaches serve as valuable baselines or references for future researchendeavors.

MenatQA: A New Dataset for Testing the Temporal Comprehension and Reasoning Abilities of Large Language Models