Reasoning, Code, or Both? How Large Language Models Handle Variations in Math Questions
A recent study evaluates how Large Language Models (LLMs) perform on mathematical reasoning tasks when faced with variations in problem statements. The research compares three methods: chain-of-thought prompting, single-shot code execution, and iterative code execution. Results indicate that while all methods showed some accuracy drop, chain-of-thought prompting was the most robust against variations.
- ▪Large Language Models achieve high accuracy on math benchmarks but struggle with variations in problem statements.
- ▪The study tested three approaches on 1,000 problems from the GSM-Symbolic dataset.
- ▪Chain-of-thought prompting was found to be the most robust method, with the least accuracy drop.
arXiv cs.AI files mainly under ai research. We currently carry 1,128 of its stories.
Story provenance
Source · retrieval · rights · ranking — open for full record
inspect →
Story provenance
Attribution is not the same as permission. This drawer separates discovery metadata, excerpts, WeSearch-generated summaries, reuse status, and whether the publisher receives the visit. Nothing here claims a legal grant the publisher has not made.
Record
| Original publisher | arXiv cs.AI |
| Canonical URL | https://arxiv.org/abs/2605.26414 |
| Publication time | Wed, 27 May 2026 00:00:00 -0400 |
| Retrieval time | 2026-05-27T04:07:56.398Z |
| Last seen | 2026-05-27T04:07:56.398Z |
| Headline source | Publisher (no WeSearch rewrite) |
| Excerpt source | publisher body |
| Excerpt method | First ~120 words (~800 chars) of extracted publisher body, fair-use limited. |
| Summary | WeSearch · cerebras-chat (WeSearch summarizer) |
| Summary source text | contentText |
| Citation coverage | Summary is a WeSearch-generated derivative; primary citation is the original publisher URL. |
| Cluster | LTFIvDy4o-NO |
| Cluster logic | Grouped by semantic title/content similarity across sources within a rolling window. Same-publisher template collisions are excluded from coverage comparison. |
| Ranking reason | Story pages are not engagement-ranked. Hub feeds use recency, with optional source-diversified chronological ordering (cap consecutive stories per source). No personalized ranking. |
| Publisher visit | Yes — open original |
| Substitutes article? | No — link-out required for full text |
Rights status (four layers)
WeSearch handling by dimension
| Indexing | May the item be indexed (stored, ranked, made findable)? | Allowed |
| Snippet | May a short excerpt of the publisher's text be shown? | Allowed |
| AI summary | May WeSearch generate its own short summary of the article? | Limited |
| Retrieval / RAG | May the content be exposed for third-party retrieval-augmented generation? | Not asserted |
| Model training | May the content be used to train AI models? | Not asserted |
| Commercial reuse | May the content be reused commercially? | Not permitted |
Basis: Derived from the published RSS/Atom feed. Contact: [email protected]. Reviewed: 2026-07-24.
Opening excerpt (first ~120 words) tap to expand
Computer Science > Artificial Intelligence arXiv:2605.26414 (cs) [Submitted on 26 May 2026] Title:Reasoning, Code, or Both? How Large Language Models Handle Variations in Math Questions Authors:Matthew Kutakh View a PDF of the paper titled Reasoning, Code, or Both? How Large Language Models Handle Variations in Math Questions, by Matthew Kutakh View PDF HTML (experimental) Abstract:Large Language Models (LLMs) achieve impressive accuracy on mathematical reasoning benchmarks, yet their performance drops when problems are modified with simple changes like different names or numbers.
…
Excerpt limited to ~120 words for fair-use compliance. The full article is at arXiv cs.AI.
Discussion
0 commentsMore from arXiv cs.AI
