Last call: ImageEval 2026 shared task at ArabicNLP

[hunzalah]

Assalamoalikum, Hello SIGARAB,

A final call to join ImageEval 2026: Cultural Grounding in Arabic Multimodal Generation and Understanding, a shared task at ArabicNLP 2026. Individual participants and teams are both welcome.

Task 1, AynVQA. 18 MENA countries, English and MSA:

1a, spoken image QA: answer culturally grounded questions that are asked aloud about an image
1b, hallucination in image QA: Spot which statements about an image are actually grounded

Task 2, CRAI-Bench. Evaluate AI-generated Arabic text-to-image outputs across five cultural dimensions, scored against human ratings. 

Dates (AoE)
Test phase closes: 7 August 2026
System papers due: 19 August 2026

Website: https://imageeval2026.github.io/ 

Data | Baselines and scorers

Codabench 1a EN · 1a MSA · 1b EN · 1b MSA · Task 2
Slack

Colab baselines are ready to run, so a first submission takes very little setup. 

Feel free to email me for any questions.

Best regards,
The ImageEval 2026 organizers

[hunzalah]

A quick update on ImageEval 2026:

The test phase has been extended to 9 August (AoE) (10 August, 3 PM Qatar time), giving participants the full weekend to build and refine their systems.
If you haven't had a chance to participate yet, there's still time! We warmly invite you to take part in any of the shared tasks, especially Spoken Visual QA and CRAI-Bench. If you know colleagues or students who may be interested, we'd greatly appreciate it if you could share the shared task with them as well.

Website: https://imageeval2026.github.io/ 

Data | Baselines and scorers

Codabench 1a EN · 1a MSA · 1b EN · 1b MSA · Task 2
Slack

If you have any questions, feel free to reply to this email.

Best regards,
The ImageEval 2026 Organizers