# Fixing Large Language Models' Specification Misunderstanding for Better Code Generation

arXiv: https://arxiv.org/abs/2309.16120 · HTML: https://mlsanity.com/paper/2309.16120/

## Key facts

- Cited by 18 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 28 Sep 2023.

## Built on by (most cited first, 18 of 18)

- [CRUXEval: A Benchmark for Code Reasoning, Understanding and Execution](https://mlsanity.com/paper/2401.03065/) (2024)
- [Trae Agent: An LLM-based Agent for Software Engineering with Test-time Scaling](https://mlsanity.com/paper/2507.23370/) (2025)
- [Code to Think, Think to Code: A Survey on Code-Enhanced Reasoning and Reasoning-Driven Code Intelligence in LLMs](https://mlsanity.com/paper/2502.19411/) (2025)
- [Hallucination to Consensus: Multi-Agent LLMs for End-to-End JUnit Test Generation](https://mlsanity.com/paper/2506.02943/) (2025)
- [Large Language Models for Software Engineering: A Reproducibility Crisis](https://mlsanity.com/paper/2512.00651/) (2025)
- [RefAgent: A Multi-agent LLM-based Framework for Automatic Software Refactoring](https://mlsanity.com/paper/2511.03153/) (2025)
- [Planning-Driven Programming: A Large Language Model Programming Workflow](https://mlsanity.com/paper/2411.14503/) (2024)
- [Reasoning as a Resource: Optimizing Fast and Slow Thinking in Code Generation Models](https://mlsanity.com/paper/2506.09396/) (2025)
- [Self-Explained Keywords Empower Large Language Models for Code Generation](https://mlsanity.com/paper/2410.15966/) (2024)
- [NES: An Instruction-Free, Low-Latency Next Edit Suggestion Framework Powered by Learned Historical Editing Trajectories](https://mlsanity.com/paper/2508.02473/) (2025)
- [TENET: One Step Toward Test-Driven Development for Repository-Level Code Generation](https://mlsanity.com/paper/2509.24148/) (2025)
- [Defective Task Descriptions in LLM-Based Code Generation: Detection and Analysis](https://mlsanity.com/paper/2604.24703/) (2026)
- [A Blueprint for AI-Driven Software Quality: Integrating LLMs with Established Standards](https://mlsanity.com/paper/2505.13766/) (2025)
- [Knowledge-Graph-Driven Data Synthesis for Low-Resource Software Development: A HarmonyOS Case Study](https://mlsanity.com/paper/2512.00380/) (2025)
- [From Helpful to Trustworthy: LLM Agents for Pair Programming](https://mlsanity.com/paper/2604.10300/) (2026)
- [Improving LLM Code Generation via Requirement-Aware Curriculum Reinforcement Learning](https://mlsanity.com/paper/2605.00433/) (2026)
- [Agent-Based Test Assertion Generation via Diverse Perspective Aggregation](https://mlsanity.com/paper/2608.05822/) (2026)
- [Security in LLM-Assisted Coding: A Systematic Literature Review](https://mlsanity.com/paper/2610.15482/) (2026)

## Abstract (arXiv)

Code generation is to automatically generate source code conforming to a given programming specification, which has received extensive attention especially with the development of large language models (LLMs). Due to the inherent difficulty of code generation, the code generated by LLMs may not be aligned with the specification. Although thought-eliciting prompting techniques have been proposed to enhance the code generation performance of LLMs, producing correct understanding for complicated programming problems remains challenging, resulting in unsatisfactory performance. Also, some feedback-based prompting techniques have been proposed to fix incorrect code using error messages produced by test execution. However, when the generated code deviates significantly from the ground truth, they encounter difficulties in improving performance based on such coarse-grained information. In this work, we propose a novel prompting technique, called μFiX, to improve the code generation performance of LLMs by devising both sophisticated thought-eliciting prompting and feedback-based prompting and making the first exploration on their synergy. It first exploits test case analysis to obtain specification understanding and enables a self-improvement process to identify and refine the misunderstanding in the thought-eliciting prompting phase. μFiX further fixes the specification understanding towards the direction reducing the gap between the provided understanding (from the first phase) and the actual understanding implicitly utilized by LLMs for code generation in the feedback-based prompting phase. By improving the understanding with μFiX, the code generation performance of LLMs can be largely improved. Our evaluation on two advanced LLMs (ChatGPT and DeepSeek-Coder) with six widely-used benchmarks by comparing with 15 baselines, demonstrates the effectiveness of μFiX ...