# Do Language Models Reason Across Languages?

arXiv: https://arxiv.org/abs/2601.06644 · HTML: https://mlsanity.com/paper/2601.06644/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Placed in Reasoning benchmarks and evaluation, one of 17 directions in LLM reasoning.
- Builds on 9 of its 37 references that are in the map.
- First posted on arXiv 10 Jan 2026.

## Placed in

- LLM reasoning / [Reasoning benchmarks and evaluation](https://mlsanity.com/lr/reasoning-benchmarks-and-evaluation/)

## Builds on (most cited first, 9 of 9)

- [Chain-of-Thought Prompting Elicits Reasoning in Large Language Models](https://mlsanity.com/paper/2201.11903/) (2022)
- [Large Language Models are Zero-Shot Reasoners](https://mlsanity.com/paper/2205.11916/) (2022)
- [Gemma 3 Technical Report](https://mlsanity.com/paper/2503.19786/) (2025)
- [Understanding Neural Networks through Representation Erasure](https://mlsanity.com/paper/1612.08220/) (2016)
- [Chain-of-Thought Reasoning Without Prompting](https://mlsanity.com/paper/2402.10200/) (2024)
- [NoLiMa: Long-Context Evaluation Beyond Literal Matching](https://mlsanity.com/paper/2502.05167/) (2025)
- [Logical Reasoning in Large Language Models: A Survey](https://mlsanity.com/paper/2502.09100/) (2025)
- [How Do LLMs Perform Two-Hop Reasoning in Context?](https://mlsanity.com/paper/2502.13913/) (2025)
- [Crosslingual Capabilities and Knowledge Barriers in Multilingual Large Language Models](https://mlsanity.com/paper/2406.16135/) (2024)

## Built on by (most cited first, 1 of 1)

- [XHotpotQA: A Benchmark for Cross-Lingual Knowledge Composition in Multi-Hop Question Answering](https://mlsanity.com/paper/2608.27481/) (2026)

## Abstract (arXiv)

The real-world information sources are inherently multilingual, which naturally raises a question about whether language models can synthesize information across languages. In this paper, we introduce a simple two-hop question answering setting, where answering a question requires making inferences over two multilingual documents. We find that language models are more sensitive to language variation in answer-span documents than in those providing bridging information, despite the equal importance of both documents for answering a question. Under a step-by-step sub-question evaluation, we further show that in up to 33% of multilingual cases, models fail to infer the bridging information in the first step yet still answer the overall question correctly. This indicates that reasoning in language models, especially in multilingual settings, does not follow a faithful step-by-step decomposition. Subsequently, we show that the absence of reasoning decomposition leads to around 18% composition failure, where both sub-questions are answered correctly but fail for the final two-hop questions. To mitigate this, we propose a simple three-stage SUBQ prompting method to guide the multi-step reasoning with sub-questions, which boosts accuracy from 10.1% to 66.5%.