PatternEval
Explore score history ↗PatternEval is a diagnostic benchmark with 2,415 multimodal prompts testing four response-pattern failures: chain-of-thought leakage, repetition, contradiction, and performative reasoning.
- Released
- 2026-08-13
- Institutions
- Tencent (China) · University of Electronic Science and Technology of China · Hong Kong University of Science and Technology
- Evaluates
- Factuality, General AI, Multimodal, Reasoning
- Openness
- unknown
- Importer
- Claire Radar
- Review status
- ai-reviewed
PatternEval paper
No reported scores are on record for this benchmark yet.