# Generative Compilation: On-the-Fly Compiler Feedback as AI Generates Code

arXiv: https://arxiv.org/abs/2607.13921 · HTML: https://mlsanity.com/paper/2607.13921/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 28 of its 58 references that are in the map.
- First posted on arXiv 15 Jul 2026.

## Builds on (most cited first, 28 of 28)

- [Lost in the Middle: How Language Models Use Long Contexts](https://mlsanity.com/paper/2307.03172/) (2023)
- [Neural Machine Translation of Rare Words with Subword Units](https://mlsanity.com/paper/1508.07909/) (2015)
- [A Survey on Large Language Models for Code Generation](https://mlsanity.com/paper/2406.00515/) (2024)
- [Asleep at the Keyboard? Assessing the Security of GitHub Copilot's Code Contributions](https://mlsanity.com/paper/2108.09293/) (2021)
- [Context Length Alone Hurts LLM Performance Despite Perfect Retrieval](https://mlsanity.com/paper/2510.05381/) (2025)
- [Guiding LLMs The Right Way: Fast, Non-Invasive Constrained Generation](https://mlsanity.com/paper/2403.06988/) (2024)
- [Grammar-Aligned Decoding](https://mlsanity.com/paper/2405.21047/) (2024)
- [Compilable Neural Code Generation with Compiler Feedback](https://mlsanity.com/paper/2203.05132/) (2022)
- [Iterative Refinement of Project-Level Code Context for Precise Code Generation with Compiler Feedback](https://mlsanity.com/paper/2403.16792/) (2024)
- [BaxBench: Can LLMs Generate Correct and Secure Backends?](https://mlsanity.com/paper/2502.11844/) (2025)
- [Type-Constrained Code Generation with Language Models](https://mlsanity.com/paper/2504.09246/) (2025)
- [IterGen: Iterative Semantic-aware Structured LLM Generation with Backtracking](https://mlsanity.com/paper/2410.07295/) (2024)
- [Fast Controlled Generation from Language Models with Adaptive Weighted Rejection Sampling](https://mlsanity.com/paper/2504.05410/) (2025)
- [Program Synthesis from Polymorphic Refinement Types](https://mlsanity.com/paper/1510.08419/) (2015)
- [CRUST-Bench: A Comprehensive Benchmark for C-to-safe-Rust Transpilation](https://mlsanity.com/paper/2504.15254/) (2025)
- [ChopChop: a Programmable Framework for Semantically Constraining the Output of Language Models](https://mlsanity.com/paper/2509.00360/) (2025)
- [Constrained Sampling for Language Models Should Be Easy: An MCMC Perspective](https://mlsanity.com/paper/2506.05754/) (2025)
- [Statically Contextualizing Large Language Models with Typed Holes](https://mlsanity.com/paper/2409.00921/) (2024)
- [Hazelnut: A Bidirectionally Typed Structure Editor Calculus](https://mlsanity.com/paper/1607.04180/) (2016)
- [Aeneas: Rust Verification by Functional Translation](https://mlsanity.com/paper/2206.07185/) (2022)
- [Constrained Decoding for Fill-in-the-Middle Code Language Models via Efficient Left and Right Quotienting of Context-Sensitive Grammars](https://mlsanity.com/paper/2402.17988/) (2024)
- [Evaluating and Improving Automated Repository-Level Rust Issue Resolution with LLM-based Agents](https://mlsanity.com/paper/2602.22764/) (2026)
- [Live Functional Programming with Typed Holes](https://mlsanity.com/paper/1805.00155/) (2018)
- [RustHorn: CHC-based Verification for Rust Programs (full version)](https://mlsanity.com/paper/2002.09002/) (2020)
- [TreeCoder: Systematic Exploration and Optimisation of Decoding and Constraints for LLM Code Generation](https://mlsanity.com/paper/2511.22277/) (2025)
- [Learning to Blame: Localizing Novice Type Errors with Data-Driven Diagnosis](https://mlsanity.com/paper/1708.07583/) (2017)
- [Oxide: The Essence of Rust](https://mlsanity.com/paper/1903.00982/) (2019)
- [Type-Error Ablation and AI Coding Agents](https://mlsanity.com/paper/2606.01522/) (2026)

## Abstract (arXiv)

Languages with rich static semantics, such as Rust, provide stronger guarantees for AI-generated code, but their strictness makes generation more difficult. Off-the-shelf compilers can provide useful feedback post-generation, but does not guide intermediate generation steps, such as those during autoregressive LLM decoding. Constrained decoding intervenes earlier by rejecting invalid tokens during sampling, but requires white-box model access and costly reimplementation for semantic constraints. We introduce generative compilation, the first approach to obtaining compiler feedback on partial programs during generation. The core technical device is a sealor: a lightweight, mostly syntax-guided transformation that converts partial programs into complete ones that standard compilers can diagnose. It is designed such that possible-to-complete partial programs are never rejected, while preserving enough code context to catch genuine dead ends early. We construct such a sealor on a core Rust-like calculus and prove that it satisfies these properties, all mechanized in Lean. We extend it to the first partial-program checker for real Rust. We evaluate our method on challenging repository-level Rust coding tasks, across both frontier black-box and open-weight models. We show that generative compilation reduces non-compiling outputs and improves functional correctness, relative to standard post-generation feedback. It does so by detecting a broad range of errors close to their source and early during generation, thereby reducing errors cascades and enabling focused diagnostics. More broadly, generative compilation is a step toward making compilers a first-class citizen of AI-assisted programming active during generation, rather than a separate post-generation check.