MetroLLM-Bench
Explore score history ↗MetroLLM-Bench evaluates language models as transit kiosk policy layers across six real metro systems. It includes 955 cases covering routing, fare calculation, disruptions, accessibility, and adversarial input, with models required to call structured tools and submit machine-renderable terminal states.
- Released
- 2026-09-09
- Evaluates
- General AI, Language & Knowledge, cs.LG
- Openness
- unknown
- Importer
- Claire Radar
- Review status
- ai-reviewed
MetroLLM-Bench paper and code
No reported scores are on record for this benchmark yet.