ElderBench
Explore score history ↗ElderBench is a benchmark for evaluating mobile GUI agents on tasks elicited from older adults, focusing on natural language patterns like indirect speech and ambiguity. It assesses agent performance on 249 tasks across 20 apps.
- Released
- 2026-09-04
- Institutions
- Fudan University
- Evaluates
- Agents & Tool Use, Mobile & Personal Computing, cs.AI
- Openness
- unknown
- Importer
- Claire Radar
- Review status
- ai-name-audit-deferred
ElderBench paper
No reported scores are on record for this benchmark yet.