Create a paper2code task (pass parameters in JSON)

Create a paper2code task from the paper: https://arxiv.org/abs/2504.17192
Corresponding open-source project: https://github.com/going-doer/Paper2Code

In the field of machine learning, code implementations for scientific papers are often missing, which slows down research reproducibility and innovation. However, recently, large language models (LLMs) have demonstrated outstanding performance in understanding scientific literature and generating high-quality code. To address this, we propose PaperCoder, a multi-agent LLM framework for automatically converting machine learning papers into functional code repositories. The PaperCoder workflow is divided into three stages: the planning stage formulates an implementation roadmap, designs the system architecture, identifies file dependencies, and generates configuration files; the analysis stage interprets implementation details; and the generation stage produces modular, dependency-aware code. Our experiments demonstrate the effectiveness of PaperCoder in high-quality implementations, especially excelling in the Paper2CodeBench and PaperBench Code-Dev benchmarks. In addition, the framework shows strong execution capabilities across different LLM models, requiring only minor modifications to 0.81% of code lines to run. Despite implementation challenges, PaperCoder represents an important step toward accelerating scientific research progress. Future work includes expanding to other domains where validation is not code-centric and enhancing the ability to handle visual inputs.

After successfully creating the task, obtain the task_id. Use the task_id to poll the task query interface. Each task typically takes about 10 minutes.

Pricing: Calculated based on the model invocation cost

Request Code Samples

Shell

JavaScript

Java

Swift

PHP

Python

HTTP

Objective-C

Ruby

OCaml

Dart

Create a paper2code task (pass parameters in JSON)

Request

Request Code Samples

Responses