# A Four-Stage Decomposition of Word-Problem Solving and Mechanistic Fragility in LLM Math Reasoning

> Заголовок на английском

> Полный текст доступен на английском

**Published:** 17 сентября 2026 г.
**Author:** Zhongdi Qu, Carla P. Gomes
**Source:** arXiv — cs.AI
**Language:** en-US
**HTML:** https://aiqazaqstan.com/ru/articles/a-four-stage-decomposition-of-word-problem-solving-and-mechanistic-fragi

Large language models solve grade-school math word problems with high accuracy, yet a single irrelevant clause inserted into the problem can collapse it. We reconcile these observations with a mechanistic account. We show that the model's internal computation decomposes into a…

> Материал агрегирован из первоисточника: **arXiv — cs.AI**
> [Читать полностью в arXiv — cs.AI](https://arxiv.org/abs/2609.17804)

**Теги:** #arXiv

---

*AI Qazaqstan — non-commercial informational portal about AI in Kazakhstan. Machine translations and summaries are AI-generated.*