What Really Controls Temporal Reasoning in Large Language Models: Tokenisation or Representation of Time?
On the map: LLM reasoning → Knowledge- and tool-augmented reasoning
Abstract (arXiv)
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: https://github.com/gagan3012/mtb
Key facts. Cited by 1 paper in the map (data through 7 Oct 2026). Placed in Knowledge- and tool-augmented reasoning, one of 17 directions in LLM reasoning. Builds on 16 of its 18 references that are in the map. First posted on arXiv 19 Mar 2026.
Builds on
16 of 18 references.
Closest ancestors · 4
Same direction, newest first
- 2026Measuring Iterative Temporal Reasoning with Time Puzzles12 Jan
- 2025Can Language Models Handle a Non-Gregorian Calendar? The Case of the Japanese wareki4 Sep
- 2025Date Fragments: A Hidden Bottleneck of Tokenization for Temporal Reasoning22 May
- 2024Test of Time: A Benchmark for Evaluating LLMs on Temporal Reasoning13 Jun
Other references · 12
- 2023Language Model Tokenizers Introduce Unfairness Between Languagescited by 207
- 2023Do All Languages Cost the Same? Tokenization in the Era of Commercial Language Modelscited by 186
- 2025FineWeb2: One Pipeline to Scale Them All -- Adapting Pre-Training Data Processing to Every Languagecited by 134
- 2024Tokenization counts: the impact of tokenization on arithmetic in frontier LLMscited by 112
- 2025ChronoSense: Exploring Temporal Understanding in Large Language Models with Time Intervals of Eventscited by 17
- 2024The Geometry of Numerical Reasoning: Language Models Compare Numeric Properties in Linear Subspacescited by 10
- 2025Lost in Time: Clock and Calendar Understanding Challenges in Multimodal LLMscited by 10
- 2025Tokenization and Representation Biases in Multilingual Models on Dialectal NLP Taskscited by 9
- 2025Around the World in 24 Hours: Probing LLM Knowledge of Time and Placecited by 6
- 2025Efficient numeracy in language models through single-token number embeddingscited by 6
- 2025Diagnosing Memorization in Chain-of-Thought Reasoning, One Token at a Timecited by 4
- 2025LLMLagBench: Identifying Temporal Training Boundaries in Large Language Modelscited by 3
Built on by
Citing papers per year: 2026: 1.
1 paper.
Similar, not cited
Close in wording, no citation either way
- 2024DateLogicQA: Benchmarking Temporal Biases in Large Language Models
- 2025TIME: A Multi-level Benchmark for Temporal Reasoning of LLMs in Real-World Scenarios
- 2025SPAN: Benchmarking and Improving Cross-Calendar Temporal Reasoning of Large Language Models
- 2023TimeBench: A Comprehensive Evaluation of Temporal Reasoning Abilities in Large Language Models
- 2024Time Awareness in Large Language Models: Benchmarking Fact Recall Across Time
- 2023Unlocking Temporal Question Answering for Large Language Models with Tailor-Made Reasoning Logic
- 2024Remember This Event That Year? Assessing Temporal Information and Reasoning in Large Language Models
- 2025Benchmarking Temporal Reasoning and Alignment Across Chinese Dynasties