# Fine-Tuning Qwen3-27B for C-to-Rust Code Translation: A Three-Stage Curriculum of Pretraining, Debugging-Aware SFT, and Task-Specific SFT

arXiv: https://arxiv.org/abs/2608.13681 · HTML: https://mlsanity.com/paper/2608.13681/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 38 of its 59 references that are in the map.
- First posted on arXiv 13 Aug 2026.

## Builds on (most cited first, 30 of 38)

- [The Llama 3 Herd of Models](https://mlsanity.com/paper/2407.21783/) (2024)
- [SWE-bench: Can Language Models Resolve Real-World GitHub Issues?](https://mlsanity.com/paper/2310.06770/) (2023)
- [Dream 7B: Diffusion Large Language Models](https://mlsanity.com/paper/2508.15487/) (2025)
- [LLaDA 1.5: Variance-Reduced Preference Optimization for Large Language Diffusion Models](https://mlsanity.com/paper/2505.19223/) (2025)
- [Magicoder: Empowering Code Generation with OSS-Instruct](https://mlsanity.com/paper/2312.02120/) (2023)
- [Mercury: Ultra-Fast Language Models Based on Diffusion](https://mlsanity.com/paper/2506.17298/) (2025)
- [CodeNet: A Large-Scale AI for Code Dataset for Learning a Diversity of Coding Tasks](https://mlsanity.com/paper/2105.12655/) (2021)
- [DiffuCoder: Understanding and Improving Masked Diffusion Models for Code Generation](https://mlsanity.com/paper/2506.20639/) (2025)
- [Qwen3.5-Omni Technical Report](https://mlsanity.com/paper/2604.15804/) (2026)
- [Dream-Coder 7B: An Open Diffusion Language Model for Code](https://mlsanity.com/paper/2509.01142/) (2025)
- [A Survey on Diffusion Language Models](https://mlsanity.com/paper/2508.10875/) (2025)
- [Efficient Reasoning with Hidden Thinking](https://mlsanity.com/paper/2501.19201/) (2025)
- [Verus: Verifying Rust Programs using Linear Ghost Types (extended version)](https://mlsanity.com/paper/2303.05491/) (2023)
- [Token Reduction Should Go Beyond Efficiency in Generative Models -- From Vision, Language to Multimodality](https://mlsanity.com/paper/2505.18227/) (2025)
- [Code Translation with Compiler Representations](https://mlsanity.com/paper/2207.03578/) (2022)
- [LazyDiT: Lazy Learning for the Acceleration of Diffusion Transformers](https://mlsanity.com/paper/2412.12444/) (2024)
- [Automated Proof Generation for Rust Code via Self-Evolution](https://mlsanity.com/paper/2410.15756/) (2024)
- [CodeFusion: A Pre-trained Diffusion Model for Code Generation](https://mlsanity.com/paper/2310.17680/) (2023)
- [AlphaTrans: A Neuro-Symbolic Compositional Approach for Repository-Level Code Translation and Validation](https://mlsanity.com/paper/2410.24117/) (2024)
- [Numerical Pruning for Efficient Autoregressive Models](https://mlsanity.com/paper/2412.12441/) (2024)
- [DraftAttention: Fast Video Diffusion via Low-Resolution Attention Guidance](https://mlsanity.com/paper/2505.14708/) (2025)
- [Pruning Foundation Models for High Accuracy without Retraining](https://mlsanity.com/paper/2410.15567/) (2024)
- [CRUST-Bench: A Comprehensive Benchmark for C-to-safe-Rust Transpilation](https://mlsanity.com/paper/2504.15254/) (2025)
- [EvoC2Rust: A Skeleton-guided Framework for Project-Level C-to-Rust Translation](https://mlsanity.com/paper/2508.04295/) (2025)
- [C2SaferRust: Transforming C Projects into Safer Rust with NeuroSymbolic Techniques](https://mlsanity.com/paper/2501.14257/) (2025)
- [Rethinking Token Reduction for State Space Models](https://mlsanity.com/paper/2410.14725/) (2024)
- [Search for Efficient Large Language Models](https://mlsanity.com/paper/2409.17372/) (2024)
- [FastCar: Cache Attentive Replay for Fast Auto-Regressive Video Generation on the Edge](https://mlsanity.com/paper/2505.14709/) (2025)
- [SACTOR: LLM-Driven Correct and Idiomatic C to Rust Translation with Static Analysis and FFI-Based Verification](https://mlsanity.com/paper/2503.12511/) (2025)
- [Translating C To Rust: Lessons from a User Study](https://mlsanity.com/paper/2411.14174/) (2024)

## Abstract (arXiv)

Translating C code into safe, idiomatic Rust is a longstanding software-engineering goal because it can eliminate entire classes of memory-safety vulnerabilities while preserving the functional behavior of legacy systems. Large language models (LLMs) have shown promise for this task but typically underperform when applied off-the-shelf, since general-purpose pretraining rarely emphasizes idiomatic Rust generation, cross-language semantic equivalence, or the ability to reason about and repair compiler/runtime feedback. In this report we describe a three-stage fine-tuning curriculum applied to Qwen3-27B that is designed to progressively specialize the model for the C-to-Rust (C2Rust) translation task: (1) continued pretraining on Rust-centric corpora to strengthen the model's prior over idiomatic Rust syntax and standard-library usage; (2) supervised fine-tuning (SFT) on the microsoft/Verus_Training_Data dataset to instill debugging and self-repair behavior over Rust code; and (3) task-specific SFT on paired C/Rust solutions derived from LeetCode problems to teach direct semantic translation. We evaluate the resulting model using the agentic, static-analysis-guided verification framework of SACTOR, which performs structure-aware, two-phase (unidiomatic to idiomatic) translation with foreign-function-interface (FFI)-based end-to-end (E2E) testing. We report success rate, idiomaticity (Clippy lint counts, unsafe-code fraction), and failure-mode analyses, and compare our fine-tuned model against baseline Qwen3-27B and other LLMs evaluated under the same framework.