AI Paper Atlas

Research map / Large language models / Agents, code and math

RPG: A Repository Planning Graph for Unified and Scalable Codebase Generation

Jane Luo, Xin Zhang, Steven Liu, Jie Wu, Jianfeng Liu, Yiming Huang, Yangyu Huang, Chengyu Yin, Ying Xin, Yuefeng Zhan, Hao Sun, Qi Chen and 2 more

ICLR 2026 · Large language models › Agents, code and math · approach: Code generation benchmarks · 18 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
use RPG-structured planning instead of unstructured planning (both without EFT) (ablation)ZeroRepo · RepoCraft MLKit-Pycoverage (%): 59.6 -> 87.2Improved
use RPG-structured planning instead of unstructured planning (both without EFT) (ablation)ZeroRepo · RepoCraft MLKit-Pylines of code: 25,202 LOC; 7x volume increaseImproved
add the EpiCoder Feature Tree knowledge base (ablation)ZeroRepo (o3-mini) · RepoCraft MLKit-Pycoverage (%) speed: 97.9% by Iteration 40 without EFT matches EFT version at Iteration 30Improved
add the EpiCoder Feature Tree to LOC growth (ablation)ZeroRepo (o3-mini) · RepoCraft MLKit-PyLOC vs iteration fit: y ≈ 983x + 2992 vs y ≈ 800x + 989Improved
replace Claude Code with ZeroRepo (o3-mini) as agentClaude Code CLI · RepoCraftcoverage (%): 54.2 -> 81.5Improved
replace Claude Code with ZeroRepo (o3-mini) as agentClaude Code CLI · RepoCraftpass rate (%): 33.9 -> 69.7Improved
use Qwen3-Coder backbone instead of o3-mini for ZeroRepoZeroRepo · RepoCraftcoverage / pass rate / LOC: 81.5 -> 75.1 coverage; 69.7 -> 57.3 pass rate; 23,977.3 -> 36,941.0 LOCMixed
increase planning iterations from 5 to 30ZeroRepo · RepoCraftcoverage (%): 70.2 -> nearly 96Improved
add graph-guided search (RPG) to agent localization (ablation)coding agent · Integration testing, incremental development, debugginglocalization steps: reduces effort by 30-50%Improved
use o3-mini as automated evaluator versus humanZeroRepo evaluation · RepoCraftPearson correlation with human: 0.893 coverage and 0.958 noveltyImproved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy