You Don't Need Prompt Engineering Anymore: The Prompting Inversion
On the map: LLM reasoning → Prompting and chain-of-thought reasoning
Abstract (arXiv)
Prompt engineering, particularly Chain-of-Thought (CoT) prompting, significantly enhances LLM reasoning capabilities. We introduce "Sculpting," a constrained, rule-based prompting method designed to improve upon standard CoT by reducing errors from semantic ambiguity and flawed common sense. We evaluate three prompting strategies (Zero Shot, standard CoT, and Sculpting) across three OpenAI model generations (gpt-4o-mini, gpt-4o, gpt-5) using the GSM8K mathematical reasoning benchmark (1,317 problems). Our findings reveal a "Prompting Inversion": Sculpting provides advantages on gpt-4o (97% vs. 93% for standard CoT), but becomes detrimental on gpt-5 (94.00% vs. 96.36% for CoT on full benchmark). We trace this to a "Guardrail-to-Handcuff" transition where constraints preventing common-sense errors in mid-tier models induce hyper-literalism in advanced models. Our detailed error analysis demonstrates that optimal prompting strategies must co-evolve with model capabilities, suggesting simpler prompts for more capable models.
Key facts. Cited by 2 papers in the map (data through 7 Oct 2026). Placed in Prompting and chain-of-thought reasoning, one of 17 directions in LLM reasoning. Builds on 16 of its 18 references that are in the map. First posted on arXiv 25 Oct 2025.
Builds on
16 of 18 references.
Closest ancestors · 6
Same direction, newest first
- 2023Graph of Thoughts: Solving Elaborate Problems with Large Language Models18 Aug
- 2023Plan-and-Solve Prompting: Improving Zero-Shot Chain-of-Thought Reasoning by Large Language Models6 May
- 2022Large Language Models are Zero-Shot Reasoners24 May
- 2022Least-to-Most Prompting Enables Complex Reasoning in Large Language Models21 May
- 2022Self-Consistency Improves Chain of Thought Reasoning in Language Models21 Mar
- 2022Chain-of-Thought Prompting Elicits Reasoning in Large Language Models28 Jan
Other references · 10
- 2020Language Models are Few-Shot Learnerscited by 32,199
- 2023GPT-4 Technical Reportcited by 18,132
- 2022Training language models to follow instructions with human feedbackcited by 15,712
- 2021Training Verifiers to Solve Math Word Problemscited by 9,743
- 2020Scaling Laws for Neural Language Modelscited by 6,411
- 2023Tree of Thoughts: Deliberate Problem Solving with Large Language Modelscited by 3,155
- 2022Solving Quantitative Reasoning Problems with Language Modelscited by 1,359
- 2022PAL: Program-aided Language Modelscited by 635
- 2022Decomposed Prompting: A Modular Approach for Solving Complex Taskscited by 554
- 2025From Literal to Liberal: A Meta-Prompting Framework for Eliciting Human-Aligned Exception Handling in Large Language Modelscited by 1
Built on by
Citing papers per year: 2026: 2.
2 papers.
Similar, not cited
Close in wording, no citation either way
- 2025Why Prompt Design Matters and Works: A Complexity Analysis of Prompt Search Space in LLMs
- 2022Towards Understanding Chain-of-Thought Prompting: An Empirical Study of What Matters
- 2025Mapping the Minds of LLMs: A Graph-Based Analysis of Reasoning LLM
- 2024Unveiling the Statistical Foundations of Chain-of-Thought Prompting Methods
- 2024LLMs as Method Actors: A Model for Prompt Engineering and Architecture
- 2024Towards Generalist Prompting for Large Language Models by Mental Models
- 2025Understanding LLM Scientific Reasoning through Promptings and Model's Explanation on the Answers
- 2026Soft Guidance Starts to Outperform CoT Prompting as LLMs Improve