DisasterM3-Reason
A multimodal reasoning benchmark focused on disaster scenarios, requiring agents to combine satellite imagery, geographic information and socioeconomic data to perform impact assessment and causal analysis of disaster events.
Timeline (last 90 days)
On the two benchmarks Earth-Bench and DisasterM3-Reason, OSGR with GPT-4o raised accuracy from 49.3% to 63.7%.
On the two benchmarks Earth-Bench and DisasterM3-Reason, OSGR with GPT-4o reduced the execution step ratio from 2.9 to 2.5.
DisasterM3-Reason is a multimodal reasoning benchmark focused on disaster scenarios, requiring agents to combine satellite imagery, geographic information and socioeconomic data to conduct impact assessment and causal analysis of disaster events.
All Facts (3)
On the two benchmarks Earth-Bench and DisasterM3-Reason, OSGR with GPT-4o raised accuracy from 49.3% to 63.7%.
50%UnverifiedOn the two benchmarks Earth-Bench and DisasterM3-Reason, OSGR with GPT-4o reduced the execution step ratio from 2.9 to 2.5.
50%UnverifiedDisasterM3-Reason is a multimodal reasoning benchmark focused on disaster scenarios, requiring agents to combine satellite imagery, geographic information and socioeconomic data to conduct impact assessment and causal analysis of disaster events.
50%