AI4Fire: Evaluating Large Language Models on Wildfire Tasks

Key points

• How do they perform on wildfire tasks, with and without grounding?1
• Bare means a model receives the task input alone.1
• Grounded means it also receives one task-specific addition: for smoke detection, a smoke-free reference frame from the same camera.1
• AI4Fire runs six core models bare and grounded on five wildfire tasks, zero-shot; a sweep adds 29 more.1

Sources (1)

• [1]AI4Fire: Evaluating Large Language Models on Wildfire Tasks
  arXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 7, 09:52 PM

  Large language models (LLMs) are entering wildfire management, where overstated evaluations can cost property and lives.

  How do they perform on wildfire tasks, with and without grounding?

Extractive summary: sentences quoted from the sources.