arXiv:2607. 05992v1 Announce Type: cross Abstract: Mathematical reasoning has become a central task for evaluating and tuning reasoning Large Language Models (LLMs), yet existing benchmarks remain heavily biased toward high-resource languages, with English and Chinese dominating both pre-training corpora and evaluation suites.
The evaluation of mathematical reasoning in large language models (LLMs) has predominantly focused on high-resource languages like English. This has created a significant barrier to the equitable development and deployment of AI in linguistically diverse regions such as Bangladesh, where over 230 million people speak Bengali.
The paper re‑examines the GSM‑Symbolic benchmark, arguing that its claim of widespread reasoning failures in 25 LLMs is based on weak statistics. Using bootstrapped Generalised Linear Mixed Models on 20 open‑weight models, only eight show significant performance changes, and a systematic shift toward larger integers in the dataset explains many of these effects. The authors also uncover model‑specific failure modes such as variable binding fragility, arithmetic limits, and dual‑task interference, cautioning against blanket conclusions about LLM reasoning.
arXiv:2501. 11790v5 Announce Type: replace-cross Abstract: Recent studies have raised significant concerns regarding the reliability of current mathematics benchmarks, highlighting issues such as simplistic design and potential data contamination.
arXiv:2607. 10114v1 Announce Type: cross Abstract: Reasoning Language Models (RLMs) achieve their strongest performance when they reason in English, the language for which reasoning-oriented training data is most abundant.
arXiv:2609.10445v1 Announce Type: new Abstract: Reasoning language models have made substantial advances on a variety of complex tasks, yet their capabilities remain overwhelmingly English-centric: m...
arXiv:2609.15145v1 Announce Type: new Abstract: The reasoning ability of large language models (LLMs) is a critical factor for practical LLM-based applications. To investigate the current reasoning c...
The study investigates the performance gap between native-language reasoning and English-pivoted reasoning in large language models. By creating extensive multilingual reasoning datasets and fine‑tuning specialists on Qwen/Qwen3-8B-Base, the authors find that the native reasoning gap is much smaller (1.9–3.5%) than previously reported. They analyze weight‑space changes, discover a language‑agnostic reasoning core in the middle layers, and propose a Layer Swap technique that transfers these mid‑layer updates from an English specialist to native specialists, effectively closing most of the gap while maintaining native chain‑of‑thought output.
arXiv:2606. 15080v1 Announce Type: cross Abstract: While Large Reasoning Models (LRMs) show strong performance in English, they often fail to reason in the language of the query, a phenomenon known as language collapse.
arXiv:2606. 03606v1 Announce Type: cross Abstract: Large language models achieve strong performance on arithmetic reasoning benchmarks, and one common response to arithmetic brittleness is to delegate computation to code.
arXiv:2607. 06327v1 Announce Type: cross Abstract: Uncertainty estimation (UE) enables LLM-powered systems to recognize when to abstain, yet existing research has predominantly focused on English.
Large language models achieve strong performance on arithmetic reasoning benchmarks, and one common response to arithmetic brittleness is to delegate computation to code. Yet models are still often used in settings where they must reason directly from natural language, and trustworthy models should solve small-number arithmetic word problems without external tools.