Reasoning
KRIS-Bench
Released image-editing attempts and their individual consistency, instruction-following, quality and knowledge-plausibility ratings.
8,941items
20subjects
HF task bank: CC-BY-4.0. Original GitHub release: Apache-2.0. No separate license declaration was found in the captured result repository.license
generaldomain
reasoningdomain
imagemodality
textmodality
item-level responses released
Saturation status: No
Response matrix
Loading response matrix…
lowhighUnobserved
Scale: Grades: 1, 2, 3, 4, 5