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

• URL: http://arxiv.org/abs/2608.13681v1
• Date: Thu, 13 Aug 2026 18:24:02 GMT
• Title: Fine-Tuning Qwen3-27B for C-to-Rust Code Translation: A Three-Stage Curriculum of Pretraining, Debugging-Aware SFT, and Task-Specific SFT
• Abstract summary: This report describes a three-stage fine-tuning curriculum applied to Qwen3-27B.<n>It is designed to progressively specialize the model for the C-to-Rust (C2Rust) translation task.<n>We evaluate the resulting model using the agentic, static-analysis-guided verification framework of SACTOR.
• Score: 32.52081475702783
• License: http://creativecommons.org/licenses/by-sa/4.0/
• Abstract: 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.

Related papers

• From C to Idiomatic Rust: A Ship-of-Theseus Agentic Translation [0.13537117504260618]
  C underpins operating systems, embedded platforms, and network infrastructure because its abstractions map directly to machine behaviour.<n> Rust eliminates most of failure modes through a static ownership and borrowing model that enforces memory safety and aliasing constraints at compile time.<n>This paper presents a migration methodology that first generates a semantics-preserving, non-idiomatic Rust baseline and then incrementally rewrites it into idiomatic Rust using agentic AI.
  arXiv  Detail & Related papers  (2026-07-30T21:00:27Z)
• Towards Reliable C-to-Rust Translation with Rule-Guided Reasoning and Reinforcement Learning [21.45219780181232]
  Leveraging large language models for automated C-to-Rust translation has emerged as a promising direction.<n>We propose a Translation fRAmework Via rule-guided reasoning and rEinforcement Learning, namely TRAVEL, consisting of two modules.<n>The first module employs Monte Carlo Tree Search (MCTS)-based reasoning path construction guided by Rust-specific rules.<n>The second module introduces reinforcement learning that couples execution feedback with reasoning-quality signals, encouraging the model to construct reasoning paths that accurately capture program semantics.
  arXiv  Detail & Related papers  (2026-07-22T09:50:01Z)
• Weave of Formal Thought [51.56484100374058]
  We introduce Weave of Formal Thought (WoFT), a paradigm uniting rigorous syntactic validation with learned structural representations.<n>We present a latent-variable fine-tuning method training the language model to interleave non-terminal grammar symbols directly into generation.<n>For Python, fine-tuning StarCoder2-3B with our RWS objective reduces per-token cross-entropy by 14.3% relative to a text-only SFT baseline.
  arXiv  Detail & Related papers  (2026-06-24T15:58:11Z)
• CORE: Context-Robust Remasking for Diffusion Language Models [51.59514489363897]
  We propose Context-Robust Remasking (CORE), a training-free framework for inference-time revision.<n>Rather than trusting static token probabilities, CORE identifies context-brittle tokens by probing their sensitivity to targeted masked-context perturbations.<n>On LLaDA-8B-Base, CORE delivers consistent improvements across reasoning and code benchmarks, outperforming compute-matched baselines and improving MBPP by up to 9.2 percentage points.
  arXiv  Detail & Related papers  (2026-02-04T00:12:30Z)
• Readability-Robust Code Summarization via Meta Curriculum Learning [53.44612630063336]
  In the real world, code is often poorly structured or obfuscated, significantly degrading model performance.<n>We propose RoFTCodeSum, a novel fine-tuning method that enhances the robustness of code summarization against poorly readable code.
  arXiv  Detail & Related papers  (2026-01-09T02:38:24Z)
• EvoC2Rust: A Skeleton-guided Framework for Project-Level C-to-Rust Translation [17.560908544319094]
  EvoC2Rust is an automated framework for converting complete C projects to equivalent Rust ones.<n>It employs a skeleton-guided translation strategy for project-level translation.
  arXiv  Detail & Related papers  (2025-08-06T10:31:23Z)
• CRUST-Bench: A Comprehensive Benchmark for C-to-safe-Rust Transpilation [51.18863297461463]
  CRUST-Bench is a dataset of 100 C repositories, each paired with manually-written interfaces in safe Rust as well as test cases.<n>We evaluate state-of-the-art large language models (LLMs) on this task and find that safe and idiomatic Rust generation is still a challenging problem.<n>The best performing model, OpenAI o1, is able to solve only 15 tasks in a single-shot setting.
  arXiv  Detail & Related papers  (2025-04-21T17:33:33Z)
• RustMap: Towards Project-Scale C-to-Rust Migration via Program Analysis and LLM [13.584956125542396]
  Rust offers superior memory safety while maintaining C's high performance.<n>Existing automated translation tools, such as C2Rust, may rely too much on syntactic, template-based translation.<n>This paper introduces a novel dependency-guided and large language model (LLM)-based C-to-Rust translation approach, RustMap.
  arXiv  Detail & Related papers  (2025-03-22T11:57:45Z)
• LLM-Driven Multi-step Translation from C to Rust using Static Analysis [27.122409727034192]
  Translating software written in legacy languages to modern languages, such as C to Rust, has significant benefits in improving memory safety.<n>We propose SACTOR, an LLM-driven C-to-Rust zero-shot translation tool using a two-step translation methodology.<n>SACTOR produces more natural and Rust-compliant translations compared to existing methods.
  arXiv  Detail & Related papers  (2025-03-16T14:05:26Z)
• SVTRv2: CTC Beats Encoder-Decoder Models in Scene Text Recognition [77.28814034644287]
  We propose SVTRv2, a CTC model endowed with the ability to handle text irregularities and model linguistic context.<n>We extensively evaluate SVTRv2 in both standard and recent challenging benchmarks.<n> SVTRv2 surpasses most EDTRs across the scenarios in terms of accuracy and inference speed.
  arXiv  Detail & Related papers  (2024-11-24T14:21:35Z)
• RustRepoTrans: Repository-level Code Translation Benchmark Targeting Rust [50.65321080814249]
  RustRepoTrans is the first repository-level context code translation benchmark targeting incremental translation.<n>We evaluate seven representative LLMs, analyzing their errors to assess limitations in complex translation scenarios.
  arXiv  Detail & Related papers  (2024-11-21T10:00:52Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.