SWE-bench Multimodal
Emerging4papers using it
2024first seen
The 'SWE-bench Multimodal' is a dataset/benchmark that contains multimodal data for evaluating automated program repair systems by requiring them to jointly reason over source code, textual issue descriptions, and visual artifacts like GUI screenshots.
Papers using SWE-bench Multimodal (4)
- GALA: Multimodal Graph Alignment for Bug Localization in Automated Program RepairRepoRepair: Leveraging Code Documentation for Repository-Level Automated Program RepairFailureMem: A Failure-Aware Multimodal Framework for Autonomous Software RepairSWE-bench Multimodal: Do AI Systems Generalize to Visual Software
Domains?