REVIEW 5 cited by
Unsupervised Translation of Programming Languages
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
read the original abstract
A transcompiler, also known as source-to-source translator, is a system that converts source code from a high-level programming language (such as C++ or Python) to another. Transcompilers are primarily used for interoperability, and to port codebases written in an obsolete or deprecated language (e.g. COBOL, Python 2) to a modern one. They typically rely on handcrafted rewrite rules, applied to the source code abstract syntax tree. Unfortunately, the resulting translations often lack readability, fail to respect the target language conventions, and require manual modifications in order to work properly. The overall translation process is timeconsuming and requires expertise in both the source and target languages, making code-translation projects expensive. Although neural models significantly outperform their rule-based counterparts in the context of natural language translation, their applications to transcompilation have been limited due to the scarcity of parallel data in this domain. In this paper, we propose to leverage recent approaches in unsupervised machine translation to train a fully unsupervised neural transcompiler. We train our model on source code from open source GitHub projects, and show that it can translate functions between C++, Java, and Python with high accuracy. Our method relies exclusively on monolingual source code, requires no expertise in the source or target languages, and can easily be generalized to other programming languages. We also build and release a test set composed of 852 parallel functions, along with unit tests to check the correctness of translations. We show that our model outperforms rule-based commercial baselines by a significant margin.
Forward citations
Cited by 5 Pith papers
-
Beyond Code Pairs: Dialogue-Based Data Generation for LLM Code Translation
Multi-turn dialogue traces from a Questioner-Solver LLM pipeline improve fine-tuned code translation in Fortran-to-C++ and C++-to-CUDA, with a 7B open model beating larger proprietary systems on compilation success on...
-
NKI-Agent: Domain-Specific Fine-Tuning and Agentic Tool Use for Neuron Kernel Generation
An agent with compile/verify tools reaches 77.3% NKI kernel pass rate on real Trn1 hardware with Opus 4.8, versus 6% single-shot; SFT Qwen3-30B hits 25% at ~1/100th cost, and binary-reward GRPO fails to beat SFT.
-
FlowMalTrans: Unsupervised Binary Code Translation for Malware Detection Using Flow-Adapter Architecture
FlowMalTrans uses normalizing flows to map basic blocks across ISAs, letting a single X86-64-trained malware detector handle code from six other architectures.
-
How Programming Concepts and Neurons Are Shared in Code Language Models
In Llama-based code models, programming languages are represented through an English-like intermediate token space, with language-specific neurons concentrated in bottom layers and exclusive PL neurons in top layers; ...
-
Self-Consistent Model-based Adaptation for Visual Reinforcement Learning
SCMA trains a policy-agnostic observation denoiser, using a pre-trained world model as a clean-distribution reference, and shows improved visual RL performance under distractions.
Discussion (0). Continue with ORCID to comment.