# Unlocking LLM Code Correction with Iterative Feedback Loops

arXiv: https://arxiv.org/abs/2606.17514 · HTML: https://mlsanity.com/paper/2606.17514/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 24 of its 33 references that are in the map.
- First posted on arXiv 16 Jun 2026.

## Builds on (most cited first, 24 of 24)

- [Evaluating Large Language Models Trained on Code](https://mlsanity.com/paper/2107.03374/) (2021)
- [Program Synthesis with Large Language Models](https://mlsanity.com/paper/2108.07732/) (2021)
- [The Curious Case of Neural Text Degeneration](https://mlsanity.com/paper/1904.09751/) (2019)
- [CodeGen: An Open Large Language Model for Code with Multi-Turn Program Synthesis](https://mlsanity.com/paper/2203.13474/) (2022)
- [CRITIC: Large Language Models Can Self-Correct with Tool-Interactive Critiquing](https://mlsanity.com/paper/2305.11738/) (2023)
- [A Survey on Large Language Models for Code Generation](https://mlsanity.com/paper/2406.00515/) (2024)
- [Pass@k Training for Adaptively Balancing Exploration and Exploitation of Large Reasoning Models](https://mlsanity.com/paper/2508.10751/) (2025)
- [Debug like a Human: A Large Language Model Debugger via Verifying Runtime Execution Step-by-step](https://mlsanity.com/paper/2402.16906/) (2024)
- [Pass@K Policy Optimization: Solving Harder Reinforcement Learning Problems](https://mlsanity.com/paper/2505.15201/) (2025)
- [Are Your LLMs Capable of Stable Reasoning?](https://mlsanity.com/paper/2412.13147/) (2024)
- [A Performance Study of LLM-Generated Code on Leetcode](https://mlsanity.com/paper/2407.21579/) (2024)
- [SelfEvolve: A Code Evolution Framework via Large Language Models](https://mlsanity.com/paper/2306.02907/) (2023)
- [CodeIt: Self-Improving Language Models with Prioritized Hindsight Replay](https://mlsanity.com/paper/2402.04858/) (2024)
- [GPU Kernel Scientist: An LLM-Driven Framework for Iterative Kernel Optimization](https://mlsanity.com/paper/2506.20807/) (2025)
- [On Evaluating the Efficiency of Source Code Generated by LLMs](https://mlsanity.com/paper/2404.06041/) (2024)
- [Reasoning Through Execution: Unifying Process and Outcome Rewards for Code Generation](https://mlsanity.com/paper/2412.15118/) (2024)
- [Proving the Coding Interview: A Benchmark for Formally Verified Code Generation](https://mlsanity.com/paper/2502.05714/) (2025)
- [Piloting Copilot, Codex, and StarCoder2: Hot Temperature, Cold Prompts, or Black Magic?](https://mlsanity.com/paper/2210.14699/) (2022)
- [GPT-Fathom: Benchmarking Large Language Models to Decipher the Evolutionary Path towards GPT-4 and Beyond](https://mlsanity.com/paper/2309.16583/) (2023)
- [ChatCoder: Chat-based Refine Requirement Improves LLMs' Code Generation](https://mlsanity.com/paper/2311.00272/) (2023)
- [VeriMind: Agentic LLM for Automated Verilog Generation with a Novel Evaluation Metric](https://mlsanity.com/paper/2503.16514/) (2025)
- [Security Degradation in Iterative AI Code Generation -- A Systematic Analysis of the Paradox](https://mlsanity.com/paper/2506.11022/) (2025)
- [Analyzing Prominent LLMs: An Empirical Study of Performance and Complexity in Solving LeetCode Problems](https://mlsanity.com/paper/2508.03931/) (2025)
- [Gradual Typing in an Open World](https://mlsanity.com/paper/1610.08476/) (2016)

## Built on by (most cited first, 1 of 1)

- [Recursive Self-Improvement in AI: From Bounded Self-Refinement to Autonomous Research Loops](https://mlsanity.com/paper/2607.07663/) (2026)

## Abstract (arXiv)

Large Language Models have shown remarkable capabilities in code generation. However, most existing evaluations focus only on single-attempt accuracy and overlook the iterative refinement process that is central to real-world programming. This study presents a systematic investigation of LLMs' ability to rectify their own code through execution feedback. Using real-world programming problems across four models and two major programming languages, this study evaluates performance using iterative refinement framework where LLMs receive compiler error messages and testcase feedback after each attempt. This study introduces metrics to evaluate code failures, analyze rectification patterns, and compare the effectiveness of reasoning and non-reasoning models, offering actionable insights into both the understanding and practical application of feedback loops in LLM-driven code generation systems. Results show that reasoning models consistently improve over iterations, substantially outperforming non-reasoning models in leveraging feedback, while syntactic and runtime errors are far more tractable than logical or algorithmic failures.