InfoResearchPreprintLLM-specific
DailySafe: Benchmarking Dynamic Safety Awareness in Everyday Multi-Turn Multimodal Interactions
- Published
- Record updated
Summary
DailySafe is a benchmark of 195 naturalistic multi-turn, multi-image interactions spanning 10 everyday domains and 42 task categories. It tests whether multimodal large language models can track evolving task conditions and recognize implicit contextual risks that arise as ordinary tasks unfold. The benchmark uses case-specific safety annotations and is publicly available at https://github.com/XuankunRong/DailySafe.