# Structure-Aware Fill-in-the-Middle Pretraining for Code

arXiv: https://arxiv.org/abs/2506.00204 · HTML: https://mlsanity.com/paper/2506.00204/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- Builds on 24 of its 27 references that are in the map.
- First posted on arXiv 30 May 2025.

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

- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer](https://mlsanity.com/paper/1910.10683/) (2019)
- [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)
- [SWE-bench: Can Language Models Resolve Real-World GitHub Issues?](https://mlsanity.com/paper/2310.06770/) (2023)
- [Code Llama: Open Foundation Models for Code](https://mlsanity.com/paper/2308.12950/) (2023)
- [Is Your Code Generated by ChatGPT Really Correct? Rigorous Evaluation of Large Language Models for Code Generation](https://mlsanity.com/paper/2305.01210/) (2023)
- [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)
- [SpanBERT: Improving Pre-training by Representing and Predicting Spans](https://mlsanity.com/paper/1907.10529/) (2019)
- [InCoder: A Generative Model for Code Infilling and Synthesis](https://mlsanity.com/paper/2204.05999/) (2022)
- [DataComp-LM: In search of the next generation of training sets for language models](https://mlsanity.com/paper/2406.11794/) (2024)
- [RepoBench: Benchmarking Repository-Level Code Auto-Completion Systems](https://mlsanity.com/paper/2306.03091/) (2023)
- [CRUXEval: A Benchmark for Code Reasoning, Understanding and Execution](https://mlsanity.com/paper/2401.03065/) (2024)
- [The Impact of AI on Developer Productivity: Evidence from GitHub Copilot](https://mlsanity.com/paper/2302.06590/) (2023)
- [Efficient Training of Language Models to Fill in the Middle](https://mlsanity.com/paper/2207.14255/) (2022)
- [SantaCoder: don't reach for the stars!](https://mlsanity.com/paper/2301.03988/) (2023)
- [Enabling Language Models to Fill in the Blanks](https://mlsanity.com/paper/2005.05339/) (2020)
- [Quantifying Variance in Evaluation Benchmarks](https://mlsanity.com/paper/2406.10229/) (2024)
- [AST-T5: Structure-Aware Pretraining for Code Generation and Understanding](https://mlsanity.com/paper/2401.03003/) (2024)
- [Copilot Arena: A Platform for Code LLM Evaluation in the Wild](https://mlsanity.com/paper/2502.09328/) (2025)
- [Evaluation of LLMs on Syntax-Aware Code Fill-in-the-Middle Tasks](https://mlsanity.com/paper/2403.04814/) (2024)
- [Planning-Aware Code Infilling via Horizon-Length Prediction](https://mlsanity.com/paper/2410.03103/) (2024)

## Built on by (most cited first, 4 of 4)

- [From Code Foundation Models to Agents and Applications: A Comprehensive Survey and Practical Guide to Code Intelligence](https://mlsanity.com/paper/2511.18538/) (2025)
- [Function-Aware Fill-in-the-Middle as Mid-Training for Coding Agent Foundation Models](https://mlsanity.com/paper/2607.12463/) (2026)
- [Hide to Guide: Learning via Semantic Masking](https://mlsanity.com/paper/2605.25198/) (2026)
- [Synthetic Hallucinations, Real Gains: Hard Negatives from Frontier Models for FIM Hallucination Mitigation](https://mlsanity.com/paper/2606.03130/) (2026)

## Abstract (arXiv)

Fill-in-the-Middle (FIM) is a common pretraining method for code LLMs, where models complete code segments given surrounding context. However, existing LLMs treat code as plain text and mask random character spans. We propose and evaluate AST-FIM, a pretraining strategy that leverages Abstract Syntax Trees (ASTs) to mask complete syntactic structures at scale, ensuring coherent training examples better aligned with universal code structures and common code editing patterns such as blocks, expressions, or functions. To evaluate real-world fill-in-the-middle (FIM) programming tasks, we introduce Real-FIM-Eval, a benchmark derived from 30,000+ GitHub commits across 12 languages. On infilling tasks, experiments on 1B and 8B parameter models show that AST-FIM is particularly beneficial for real-world code editing as it outperforms standard random-character FIM by up to 5 pts on standard FIM benchmarks. Our code is publicly available at https://github.com/gonglinyuan/ast_fim.