English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

In this paper, we present an LLM-based code translation method and anassociated tool called CoTran, that translates whole-programs from onehigh-level programming language to another. Current LLM-based code translationmethods lack a training approach to ensure that the translated code reliablycompiles or bears substantial functional equivalence to the input code. In ourwork, we train an LLM via reinforcement learning, by modifying the fine-tuningprocess to incorporate compiler feedback and symbolic execution (symexec)-basedequivalence testing feedback that checks for functional equivalence between theinput and output programs. The idea is to guide an LLM-in-training, viacompiler and symexec-based testing feedback, by letting it know how far it isfrom producing perfect translations. We report on extensive experimentscomparing CoTran with 14 other code translation tools that includehuman-written transpilers, LLM-based translation tools, and ChatGPT over abenchmark of more than 57,000 Java-Python equivalent pairs, and we show thatCoTran outperforms them on relevant metrics such as compilation accuracy(CompAcc) and functional equivalence accuracy (FEqAcc). For example, our toolachieves 48.68% FEqAcc, 76.98% CompAcc for Python-to-Java translation, whereasthe nearest competing tool (PLBART-base) only gets 38.26% and 75.77% resp.Also, built upon CodeT5, CoTran achieves +11.23%, +14.89% improvement on FEqAccand +4.07%, +8.14% on CompAcc for Java-to-Python and Python-to-Java translationresp.

CoTran: An LLM-based Code Translator using Reinforcement Learning with Feedback from Compiler and Symbolic Execution