Avatar

The 'Avatar' dataset is used to evaluate the accuracy and reliability of code translation between programming languages by providing a set of programs that can be compiled and tested for correctness.

Papers using Avatar (6)

Prompt Alchemy: Automatic Prompt Refinement for Enhancing Code Generation2025 · 2 citesBeyond Translation Accuracy: Addressing False Failures in LLM-Based Code Translation2026Algorithm-Based Pipeline for Reliable and Intent-Preserving Code Translation with LLMs2026NL in the Middle: Code Translation with LLMs and Intermediate Representations2025APIRAT: Integrating Multi-source API Knowledge for Enhanced Code Translation with LLMs2025AVATAR: A Parallel Corpus for Java-Python Program Translation2021 · 14 cites