What Really Controls Temporal Reasoning in Large Language Models: Tokenisation or Representation of Time?

Domain:

natural language processing

Record type:

paperdataset

Creator:

BhaIsaPey

Host:

We present MultiTempBench, a multilingual temporal reasoning benchmark spanning three tasks, date arithmetic, time zone conversion, and temporal relation extraction across five languages (English, German, Chinese, Arabic, and Hausa) and multiple calendar conventions (Gregorian, Hijri, and Chinese Lunar). MultiTempBench contains $15,000$ examples built by translating $750$ curated English questions and expanding each into controlled date-format variants. We evaluate 20 LLMs and introduce the multilingual Date Fragmentation Ratio (mDFR), calibrated with human severity ratings, together with geometric-probing analyses of internal temporal representations. We find tokenisation quality of temporal artefacts is a resource-dependent bottleneck: in low-resource languages and rarer calendar formats, fragmentation disrupts Year/Month/Day separation and accuracy collapses, while high-resource settings are often robust to digit-level splitting. Beyond tokenisation, crossed mixed-effects regression shows that temporal linearity is the strongest predictor of temporal reasoning in high-resource languages, whereas fragmentation is the stronger predictor in low-resource languages. Code is available at: github.com

Visit

arxiv.org

Languages

Hausa

Assessing Dialect Fairness and Robustness of Large Language Models in Reasoning Tasks

Language is not monolithic. While benchmarks, including those designed for multiple languages, are o

SoT: Structured-of-Thought Prompting Guides Multilingual Reasoning in Large Language Models

Recent developments have enabled Large Language Models (LLMs) to engage in complex reasoning tasks t

Large Concept Models: Language Modeling in a Sentence Representation Space

LLMs have revolutionized the field of artificial intelligence and have emerged as the de-facto tool

184 Evaluating Reasoning-Tuned Large Language Models for Clinical Decision-Making in Spine Surgery

Abstract Large language models (LLMs) such as OpenAI o1 and DeepSeek-R1 are desi

What Really Counts in Nominal Classification in Dagaare: A Mabia Language

Large Reasoning Models Are (Not Yet) Multilingual Latent Reasoners

Large reasoning models (LRMs) achieve strong performance on mathematical reasoning tasks, often attr