Research map / Large language models / Agents, code and math
SWE-Perf: Can Language Models Optimize Code Performance on Real-World Repositories?
Xinyi He, Qian Liu, Mingzhe Du, Lin Yan, ZhiJie Fan, Yiming Huang, Yin Zheng, Zejian Yuan, Zejun MA
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 changed | Where | Measured | Effect |
|---|---|---|---|
| use OpenHands agent instead of Agentless pipeline | Claude-3.7-sonnet · SWE-Perf realistic | Performance: 0.41% -> 2.26% | Improved |
| use repo-level OpenHands instead of single-pass oracle prompting | Claude-3.7-sonnet · SWE-Perf | Performance: 1.24% -> 2.26% | Improved |
| use Agentless instead of single-pass oracle prompting | Claude-3.7-sonnet · SWE-Perf | Performance: 1.24% -> 0.41% | Hurt |
| compare best LLM method to human expert patches | OpenHands · SWE-Perf | Performance: 2.26% -> 10.85% | Hurt |
| increase the number of target functions | expert and OpenHands · SWE-Perf realistic | Performance | Hurt |
| increase original runtime of the test | expert vs models · SWE-Perf | Performance | Mixed |
| restrict performance metric to correct examples only | OpenHands vs alternatives · SWE-Perf | Performance_pass | Improved |
| compare OpenHands to expert on sklearn | OpenHands · SWE-Perf sklearn | Performance: +0.4% | Improved |
Closest papers by meaning
- SWE-fficiency: Can Language Models Optimize Real-World Repositories on Real Workloads? (ICML 2026 · similarity 0.93)
- SWE-bench Goes Live! (NeurIPS 2025 · similarity 0.91)
- SWE-bench: Can Language Models Resolve Real-world Github Issues? (ICLR 2024 · similarity 0.91)
- LiveCodeBench: Holistic and Contamination Free Evaluation of Large Language Models for Code (ICLR 2025 · similarity 0.88)
- GSO: Challenging Software Optimization Tasks for Evaluating SWE-Agents (NeurIPS 2025 · similarity 0.88)
- EffiBench-X: A Multi-Language Benchmark for Measuring Efficiency of LLM-Generated Code (NeurIPS 2025 · similarity 0.88)
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.
