ContextBench

What is ContextBench?

Evaluates text-to-image models on 92 roles and 1,656 controlled prompts across context-free, related, and unrelated conditions, measuring bias persistence in…

Released

2026-08-30

Evaluates

General AI, Safety & Trustworthiness, cs.CV

Openness

unknown

Importer

Claire Radar

Review status

ai-reviewed

Reported scores

0

Source provenance

• Original evidence https://arxiv.org/abs/2608.29847

ContextBench paper, code and dataset

• Paper https://arxiv.org/abs/2608.29847
• Code https://github.com/Sina-Emami/ContextBias
• Dataset https://huggingface.co/datasets/shaghayegh/ContextBias

Which sources cite ContextBench?

• ContextBench

Related General AI, Safety & Trustworthiness, cs.CV benchmarks

• $\tau^\tau$-Bench: An Environment for End-To-End, Realistic Agent Construction
• 360CityArena
• 3DCodeBench
• 5-Dialects-BN
• A3S-Bench
• AARRI-Bench
• ActBench
• ActiveSciBench