# Thinking Before Running! Efficient Code Generation with Thorough Exploration and Optimal Refinement

arXiv: https://arxiv.org/abs/2502.17442 · HTML: https://mlsanity.com/paper/2502.17442/

## Key facts

- Cited by 3 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 23 of its 36 references that are in the map.
- First posted on arXiv 30 Dec 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, 23 of 23)

- [LoRA: Low-Rank Adaptation of Large Language Models](https://mlsanity.com/paper/2106.09685/) (2021)
- [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)
- [Qwen Technical Report](https://mlsanity.com/paper/2309.16609/) (2023)
- [Code Llama: Open Foundation Models for Code](https://mlsanity.com/paper/2308.12950/) (2023)
- [Scaling LLM Test-Time Compute Optimally can be More Effective than Scaling Model Parameters](https://mlsanity.com/paper/2408.03314/) (2024)
- [ToolLLM: Facilitating Large Language Models to Master 16000+ Real-world APIs](https://mlsanity.com/paper/2307.16789/) (2023)
- [Reinforced Self-Training (ReST) for Language Modeling](https://mlsanity.com/paper/2308.08998/) (2023)
- [OpenCodeInterpreter: Integrating Code Generation with Execution and Refinement](https://mlsanity.com/paper/2402.14658/) (2024)
- [OctoPack: Instruction Tuning Code Large Language Models](https://mlsanity.com/paper/2308.07124/) (2023)
- [MapCoder: Multi-Agent Code Generation for Competitive Problem Solving](https://mlsanity.com/paper/2405.11403/) (2024)
- [Debug like a Human: A Large Language Model Debugger via Verifying Runtime Execution Step-by-step](https://mlsanity.com/paper/2402.16906/) (2024)
- [Code Generation with AlphaCodium: From Prompt Engineering to Flow Engineering](https://mlsanity.com/paper/2401.08500/) (2024)
- [Improving Code Generation by Training with Natural Language Feedback](https://mlsanity.com/paper/2303.16749/) (2023)
- [CodeScore: Evaluating Code Generation by Learning Code Execution](https://mlsanity.com/paper/2301.09043/) (2023)
- [A Survey of Large Language Models for Code: Evolution, Benchmarking, and Future Trends](https://mlsanity.com/paper/2311.10372/) (2023)
- [AgentCoder: Multi-Agent-based Code Generation with Iterative Testing and Optimisation](https://mlsanity.com/paper/2312.13010/) (2023)
- [CodeCoT: Tackling Code Syntax Errors in CoT Reasoning for Code Generation](https://mlsanity.com/paper/2308.08784/) (2023)
- [RethinkMCTS: Refining Erroneous Thoughts in Monte Carlo Tree Search for Code Generation](https://mlsanity.com/paper/2409.09584/) (2024)
- [The Stepwise Deception: Simulating the Evolution from True News to Fake News with LLM Agents](https://mlsanity.com/paper/2410.19064/) (2024)
- [Nuances are the Key: Unlocking ChatGPT to Find Failure-Inducing Tests with Differential Prompting](https://mlsanity.com/paper/2304.11686/) (2023)
- [Text Generation: A Systematic Literature Review of Tasks, Evaluation, and Challenges](https://mlsanity.com/paper/2405.15604/) (2024)
- [Planning-Driven Programming: A Large Language Model Programming Workflow](https://mlsanity.com/paper/2411.14503/) (2024)

## Built on by (most cited first, 3 of 3)

- [The Truth Becomes Clearer Through Debate! Multi-Agent Systems with Large Language Models Unmask Fake News](https://mlsanity.com/paper/2505.08532/) (2025)
- [Breaking the Martingale Curse: Multi-Agent Debate via Asymmetric Cognitive Potential Energy](https://mlsanity.com/paper/2603.06801/) (2026)
- [CollabCoder: Plan-Code Co-Evolution via Collaborative Decision-Making for Efficient Code Generation](https://mlsanity.com/paper/2604.13946/) (2026)

## Abstract (arXiv)

Code generation is crucial in software engineering for automating the coding process efficiently. While test-time computation methods show promise, they suffer from high latency due to multiple computation rounds. To overcome this, we introduce ThinkCoder, a framework that combines thorough exploration with optimal refinement. The exploration phase diversifies the solution space by searching for potential solutions, followed by a refinement phase that enhances precision. This approach allows us to select the best solution through careful consideration before taking action, avoiding excessive trial and error. To further minimize test-time computation overhead, we introduce preference-driven optimization with Reinforced Self-Training (ReST), which uses exploration trajectories from ThinkCoder to guide LLM's evolution. This approach enhances LLM's exploration efficiency via preference learning, cutting costs while maintaining accuracy. ThinkCoder boosts the performance with a single LLM, excelling on benchmarks like HumanEval and MBPP. Compared to SOTA models, it improves Pass@1 by 3.0% over MapCoder with just 6.4% of the computation cost. Against AgentCoder, ThinkCoder achieves a 0.5% higher Pass@1 after 2 rounds, outperforming AgentCoder's 5 rounds. Additionally, ReST with success trajectories enhances efficiency, allowing models like LLaMA2-7B to achieve competitive results using only 20% of the computational resources. These results highlight the framework's effectiveness and scalability.