# Learning from Execution: Self-Evolving Memory for Private-Library Code Generation

arXiv: https://arxiv.org/abs/2604.24222 · HTML: https://mlsanity.com/paper/2604.24222/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Placed in RAG for code, one of 15 directions in Retrieval-augmented generation.
- Builds on 22 of its 27 references that are in the map.
- First posted on arXiv 27 Apr 2026.

## Placed in

- Retrieval-augmented generation / [RAG for code](https://mlsanity.com/rag/rag-for-code/)
- Synthetic data and self-improvement for LLMs / [Code and instruction data synthesis](https://mlsanity.com/synth/code-and-instruction-data-synthesis/)

## Builds on (most cited first, 22 of 22)

- [The Llama 3 Herd of Models](https://mlsanity.com/paper/2407.21783/) (2024)
- [Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks](https://mlsanity.com/paper/2005.11401/) (2020)
- [Qwen3 Technical Report](https://mlsanity.com/paper/2505.09388/) (2025)
- [Qwen2.5-Coder Technical Report](https://mlsanity.com/paper/2409.12186/) (2024)
- [DeepSeek-Coder: When the Large Language Model Meets Programming -- The Rise of Code Intelligence](https://mlsanity.com/paper/2401.14196/) (2024)
- [ChatGLM: A Family of Large Language Models from GLM-130B to GLM-4 All Tools](https://mlsanity.com/paper/2406.12793/) (2024)
- [An Empirical Study of Catastrophic Forgetting in Large Language Models During Continual Fine-tuning](https://mlsanity.com/paper/2308.08747/) (2023)
- [CodeT: Code Generation with Generated Tests](https://mlsanity.com/paper/2207.10397/) (2022)
- [Dynamic Cheatsheet: Test-Time Learning with Adaptive Memory](https://mlsanity.com/paper/2504.07952/) (2025)
- [Multi-lingual Evaluation of Code Generation Models](https://mlsanity.com/paper/2210.14868/) (2022)
- [DocPrompting: Generating Code by Retrieving the Docs](https://mlsanity.com/paper/2207.05987/) (2022)
- [Structured Chain-of-Thought Prompting for Code Generation](https://mlsanity.com/paper/2305.06599/) (2023)
- [Evo-Memory: Benchmarking LLM Agent Test-time Learning with Self-Evolving Memory](https://mlsanity.com/paper/2511.20857/) (2025)
- [Revisiting Catastrophic Forgetting in Large Language Model Tuning](https://mlsanity.com/paper/2406.04836/) (2024)
- [When Language Model Meets Private Library](https://mlsanity.com/paper/2210.17236/) (2022)
- [On the Effectiveness of Large Language Models in Domain-Specific Code Generation](https://mlsanity.com/paper/2312.01639/) (2023)
- [Exploring the Potential of Diffusion Large Language Models in Code Generation](https://mlsanity.com/paper/2509.11252/) (2025)
- [ExploraCoder: Advancing code generation for multiple unseen APIs via planning and chained exploration](https://mlsanity.com/paper/2412.05366/) (2024)
- [Private-Library-Oriented Code Generation with Large Language Models](https://mlsanity.com/paper/2307.15370/) (2023)
- [Compositional API Recommendation for Library-Oriented Code Generation](https://mlsanity.com/paper/2402.19431/) (2024)
- [AI-Driven Self-Evolving Software: A Promising Path Toward Software Automation](https://mlsanity.com/paper/2510.00591/) (2025)
- [To See is Not to Master: Teaching LLMs to Use Private Libraries for Code Generation](https://mlsanity.com/paper/2603.15159/) (2026)

## Built on by (most cited first, 2 of 2)

- [Diagnosing Knowledge Gaps in LLM Tool Use: An Agentic Benchmark for Novel API Acquisition](https://mlsanity.com/paper/2606.03657/) (2026)
- [RCL: A Retrieval-Confidence Layer for Detecting Insufficient Context in Enterprise Retrieval-Augmented Code Generation](https://mlsanity.com/paper/2609.11023/) (2026)

## Abstract (arXiv)

Large Language Models (LLMs) have achieved strong performance on general code generation, but their effectiveness drops sharply in enterprise settings where software development relies on internal private libraries absent from public pre-training corpora. Existing Retrieval-Augmented Generation (RAG) methods provide a training-free solution by retrieving static API documentation, but our analysis shows that documentation mainly helps models identify what APIs to use and remains insufficient for teaching how to use them correctly. Even with oracle API-document retrieval, LLMs still make recurring errors at the API, cross-API, and task levels, including API misuse or hallucination, flawed API composition, and incorrect solution strategies. To address this limitation, we propose MEMCoder, a training-free self-evolving memory framework for private-library code generation. MEMCoder augments existing RAG pipelines with a Multi-level Evolving Memory that continuously accumulates and reuses execution-derived Usage Guidelines at the API, cross-API, and task levels. During generation, MEMCoder retrieves both static API documentation and relevant historical memories to guide code generation; after execution, it analyzes feedback to refine memory through a closed loop of generation, execution, reflection, and update. Extensive experiments on NdonnxEval and NumbaEval show that MEMCoder consistently enhances different RAG backbones across LLMs of different scales, yielding an average absolute pass@1 improvement of 18.41 percentage points. Moreover, MEMCoder outperforms existing self-evolving memory methods and validates the effectiveness of organizing execution feedback into multi-level usage memories.