CDR-Bench
Explore score history ↗CDR-Bench is a benchmark of 3,462 tasks for evaluating large language models on faithful execution of compositional, order-sensitive data refinement recipes across four domains and 29 operators, with deterministic reference outputs for exact evaluation.
- Released
- 2026-06-30
- Institutions
- Alibaba Group (Cayman Islands) · Alibaba Group (China) · National University of Singapore
- Evaluates
- General AI, Language & Knowledge, cs.AI
- Openness
- unknown
- Importer
- Claire Radar
- Review status
- unreviewed
CDR-Bench paper
No reported scores are on record for this benchmark yet.