Reproduce “DisasterAgent: An Evidence-Driven Agent for Verifiable Disaster Report Generation From Remote Sensing Imagery” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Zhijian Li, Jiawei Song, Chao Ren, Yidan Zhang, Peijin Wang, Xian Sun, Kun Fu
DisasterAgent turns paired satellite images into disaster reports linked to evidence. It organizes tool findings across space and time in a graph, then uses image-checked claims and visual evidence to guide report writing. On DisasterM3’s captioning benchmark, it scores 2.87 out of 5 versus 1.88 for its central model. This supports higher judged report quality, not verified real-world outcomes.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.