CrossSWE benchmark shows coding agents struggle with cross-repository development
Silin Chen · hf · 2026-11-03
CrossSWE is a multi-repository benchmark from Shanghai Jiao Tong University, Xidian University and Singapore Management University that evaluates coding agents on cross-repository joint development.
- As agents achieve high success rates on single-repo SWE benchmarks, CrossSWE raises task complexity to the multi-repo interdependent scenarios common in real-world development
- Cross-repository tasks add dependency-resolution challenges; results show current LLMs struggle to maintain cross-repository consistency during joint development
More from coding & agent
- Live workshop: instrument your AI agent with OpenTelemetry and debug it via Langfuse traces — Al_Grigor · 2026-11-03
- AutoRLBench: A Benchmark Testing AI Agents on Real RL Training Failures — wzenus · 2026-11-03
- Building live TV fact-checking agents on Bedrock AgentCore at re:Invent IND421 — FlolightC · 2026-11-03
- New Tutorial: Driving Gemini for Computer Use with Cua Driver — osanseviero · 2026-11-03
- SecureInput: open-source local secret input for MCP coding agents — rishabhgusain · 2026-11-03
- Sidelight runs browser agents in your real Chrome session to dodge login walls — Equivalent_Type6726 · 2026-11-03