# NExT: Teaching Large Language Models to Reason about Code Execution

arXiv: https://arxiv.org/abs/2404.14662 · HTML: https://mlsanity.com/paper/2404.14662/

## Key facts

- Cited by 63 papers in the map (data through 14 Oct 2026).
- Placed in Reasoning for agents, planning and code, one of 17 directions in LLM reasoning.
- Builds on 65 of its 87 references that are in the map.
- First posted on arXiv 23 Apr 2024.

## Placed in

- LLM reasoning / [Reasoning for agents, planning and code](https://mlsanity.com/lr/reasoning-for-agents-planning-and-code/)

## Builds on (most cited first, 30 of 65)

- [Llama 2: Open Foundation and Fine-Tuned Chat Models](https://mlsanity.com/paper/2307.09288/) (2023)
- [Chain-of-Thought Prompting Elicits Reasoning in Large Language Models](https://mlsanity.com/paper/2201.11903/) (2022)
- [ReAct: Synergizing Reasoning and Acting in Language Models](https://mlsanity.com/paper/2210.03629/) (2022)
- [Evaluating Large Language Models Trained on Code](https://mlsanity.com/paper/2107.03374/) (2021)
- [Let's Verify Step by Step](https://mlsanity.com/paper/2305.20050/) (2023)
- [Program Synthesis with Large Language Models](https://mlsanity.com/paper/2108.07732/) (2021)
- [Self-Refine: Iterative Refinement with Self-Feedback](https://mlsanity.com/paper/2303.17651/) (2023)
- [Scaling Instruction-Finetuned Language Models](https://mlsanity.com/paper/2210.11416/) (2022)
- [Code Llama: Open Foundation Models for Code](https://mlsanity.com/paper/2308.12950/) (2023)
- [Mixtral of Experts](https://mlsanity.com/paper/2401.04088/) (2024)
- [Is Your Code Generated by ChatGPT Really Correct? Rigorous Evaluation of Large Language Models for Code Generation](https://mlsanity.com/paper/2305.01210/) (2023)
- [DeepSeek-Coder: When the Large Language Model Meets Programming -- The Rise of Code Intelligence](https://mlsanity.com/paper/2401.14196/) (2024)
- [PaLM 2 Technical Report](https://mlsanity.com/paper/2305.10403/) (2023)
- [Teaching Large Language Models to Self-Debug](https://mlsanity.com/paper/2304.05128/) (2023)
- [STaR: Bootstrapping Reasoning With Reasoning](https://mlsanity.com/paper/2203.14465/) (2022)
- [Math-Shepherd: Verify and Reinforce LLMs Step-by-step without Human Annotations](https://mlsanity.com/paper/2312.08935/) (2023)
- [Solving math word problems with process- and outcome-based feedback](https://mlsanity.com/paper/2211.14275/) (2022)
- [Show Your Work: Scratchpads for Intermediate Computation with Language Models](https://mlsanity.com/paper/2112.00114/) (2021)
- [StarCoder: may the source be with you!](https://mlsanity.com/paper/2305.06161/) (2023)
- [The Flan Collection: Designing Data and Methods for Effective Instruction Tuning](https://mlsanity.com/paper/2301.13688/) (2023)
- [Neural Turing Machines](https://mlsanity.com/paper/1410.5401/) (2014)
- [Textbooks Are All You Need](https://mlsanity.com/paper/2306.11644/) (2023)
- [Orca: Progressive Learning from Complex Explanation Traces of GPT-4](https://mlsanity.com/paper/2306.02707/) (2023)
- [Explain Yourself! Leveraging Language Models for Commonsense Reasoning](https://mlsanity.com/paper/1906.02361/) (2019)
- [Teaching Small Language Models to Reason](https://mlsanity.com/paper/2212.08410/) (2022)
- [Thinking Fast and Slow with Deep Learning and Tree Search](https://mlsanity.com/paper/1705.08439/) (2017)
- [Specializing Smaller Language Models towards Multi-Step Reasoning](https://mlsanity.com/paper/2301.12726/) (2023)
- [LEVER: Learning to Verify Language-to-Code Generation with Execution](https://mlsanity.com/paper/2302.08468/) (2023)
- [OctoPack: Instruction Tuning Code Large Language Models](https://mlsanity.com/paper/2308.07124/) (2023)
- [SPoC: Search-based Pseudocode to Code](https://mlsanity.com/paper/1906.04908/) (2019)

## Built on by (most cited first, 30 of 63)

- [Training Language Models to Self-Correct via Reinforcement Learning](https://mlsanity.com/paper/2409.12917/) (2024)
- [A Survey on Test-Time Scaling in Large Language Models: What, How, Where, and How Well?](https://mlsanity.com/paper/2503.24235/) (2025)
- [Lingma SWE-GPT: An Open Development-Process-Centric Language Model for Automated Software Improvement](https://mlsanity.com/paper/2411.00622/) (2024)
- [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)
- [Code as Agent Harness](https://mlsanity.com/paper/2605.18747/) (2026)
- [From Code to Correctness: Closing the Last Mile of Code Generation with Hierarchical Debugging](https://mlsanity.com/paper/2410.01215/) (2024)
- [Teaching Language Models to Critique via Reinforcement Learning](https://mlsanity.com/paper/2502.03492/) (2025)
- [Group-Evolving Agents: Open-Ended Self-Improvement via Experience Sharing](https://mlsanity.com/paper/2602.04837/) (2026)
- [Surveying the Effects of Quality, Diversity, and Complexity in Synthetic Data From Large Language Models](https://mlsanity.com/paper/2412.02980/) (2024)
- [Mitigating Overthinking in Large Reasoning Models via Manifold Steering](https://mlsanity.com/paper/2505.22411/) (2025)
- [What Makes Large Language Models Reason in (Multi-Turn) Code Generation?](https://mlsanity.com/paper/2410.08105/) (2024)
- [Multi-Turn Code Generation Through Single-Step Rewards](https://mlsanity.com/paper/2502.20380/) (2025)
- [PyBench: Evaluating LLM Agent on various real-world coding tasks](https://mlsanity.com/paper/2407.16732/) (2024)
- [EquiBench: Benchmarking Large Language Models' Reasoning about Program Semantics via Equivalence Checking](https://mlsanity.com/paper/2502.12466/) (2025)
- [Learning to Generate Unit Tests for Automated Debugging](https://mlsanity.com/paper/2502.01619/) (2025)
- [cAST: Enhancing Code Retrieval-Augmented Generation with Structural Chunking via Abstract Syntax Tree](https://mlsanity.com/paper/2506.15655/) (2025)
- [CodeARC: Benchmarking Reasoning Capabilities of LLM Agents for Inductive Program Synthesis](https://mlsanity.com/paper/2503.23145/) (2025)
- [From Code Foundation Models to Agents and Applications: A Comprehensive Survey and Practical Guide to Code Intelligence](https://mlsanity.com/paper/2511.18538/) (2025)
- [CodeCrash: Exposing LLM Fragility to Misleading Natural Language in Code Reasoning](https://mlsanity.com/paper/2504.14119/) (2025)
- [RepoST: Scalable Repository-Level Coding Environment Construction with Sandbox Testing](https://mlsanity.com/paper/2503.07358/) (2025)
- [What I cannot execute, I do not understand: Training and Evaluating LLMs on Program Execution Traces](https://mlsanity.com/paper/2503.05703/) (2025)
- [LLMs as Continuous Learners: Improving the Reproduction of Defective Code in Software Issues](https://mlsanity.com/paper/2411.13941/) (2024)
- [Coffee-Gym: An Environment for Evaluating and Improving Natural Language Feedback on Erroneous Code](https://mlsanity.com/paper/2409.19715/) (2024)
- [CodeNav: Beyond tool-use to using real-world codebases with LLM agents](https://mlsanity.com/paper/2406.12276/) (2024)
- [CoCoNUT: Structural Code Understanding does not fall out of a tree](https://mlsanity.com/paper/2501.16456/) (2025)
- [A Tool for In-depth Analysis of Code Execution Reasoning of Large Language Models](https://mlsanity.com/paper/2501.18482/) (2025)
- [HoarePrompt: Structural Reasoning About Program Correctness in Natural Language](https://mlsanity.com/paper/2503.19599/) (2025)
- [Agent That Debugs: Dynamic State-Guided Vulnerability Repair](https://mlsanity.com/paper/2504.07634/) (2025)
- [CodeLutra: Boosting LLM Code Generation via Preference-Guided Refinement](https://mlsanity.com/paper/2411.05199/) (2024)
- [LLMs Lean on Priors, Not Programming Language Semantics](https://mlsanity.com/paper/2510.03415/) (2025)

## Abstract (arXiv)

A fundamental skill among human developers is the ability to understand and reason about program execution. As an example, a programmer can mentally simulate code execution in natural language to debug and repair code (aka. rubber duck debugging). However, large language models (LLMs) of code are typically trained on the surface textual form of programs, thus may lack a semantic understanding of how programs execute at run-time. To address this issue, we propose NExT, a method to teach LLMs to inspect the execution traces of programs (variable states of executed lines) and reason about their run-time behavior through chain-of-thought (CoT) rationales. Specifically, NExT uses self-training to bootstrap a synthetic training set of execution-aware rationales that lead to correct task solutions (e.g., fixed programs) without laborious manual annotation. Experiments on program repair tasks based on MBPP and HumanEval demonstrate that NExT improves the fix rate of a PaLM 2 model, by 26.1% and 14.3% absolute, respectively, with significantly improved rationale quality as verified by automated metrics and human raters. Our model can also generalize to scenarios where program traces are absent at test-time.