NIJul 9

RepLLM: Toward Automatically Reproducing Network Research Results

Yining Jiang, Yunxin Xu, Wenyun Xu, Yufan Zhu, Rui Liu, Tangtang He, Haiying Huang, Letian Zhu, Qingyu Song, Qiang Su, Lizhao You, Lu Tang
arXiv:2509.2107422.71 citationsh-index: 5Has Code
Predicted impact top 1% in NI · last 90 daysOriginality Incremental advance
AI Analysis

For networking researchers, RepLLM addresses the challenge of reproducing results from academic papers, which is currently difficult due to lack of open-source code and system complexity.

RepLLM is a multi-agent framework that automates code reproduction from networking research papers, achieving 95% benchmark reproduction within two hours and reducing token consumption by up to 10% compared to baselines.

Result reproduction of computer networking research is challenging as the scarcity of open-source implementations and the complexity of heterogeneous system architectures. Even though Large Language Models have demonstrated potential in code generation, existing code generation frameworks often fail to address the long-context constraints and intricate logical dependencies, which are vital in reproducing network systems from academic papers. Thus, we introduce RepLLM, an end-to-end multi-agent framework designed to automate code reproduction from paper content. RepLLM features a collaborative architecture comprising four specialized agents -- Content Parsing, Architecture Design, Code Generation, and Audit&Repair, which are coordinated through Shared Memory mechanism to ensure global context consistency. With the enhancement of Structured Chain-of-Thought LLM reasoning and a sandbox-isolated static-dynamic debugging methodology, our framework effectively resolves semantic discrepancies and runtime errors, thereby improving reliable reproductions. Extensive evaluations on representative papers in top conferences demonstrate that RepLLM outperforms state-of-the-art system-level LLM frameworks in generating compile-ready and logically correct systems. Our results show that, with the aid of RepLLM, we can reproduce 95% of the original benchmarks within approximately two hours while reducing token consumption by up to 10% compared with state-of-the-art baselines.

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes