ROCS Benchmark

What is ROCS?

Evaluates text-to-image retrieval models on finding a named low-salience rare object in cluttered scenes, using re-captioned subsets of Flickr30K and MS COCO.

Released

2026-09-22

Evaluates

General AI, Information retrieval, Vision & 3D, cs.CV

Openness

unknown

Importer

Claire Radar

Review status

ai-reviewed

Reported scores

0

Source provenance

• Original evidence https://arxiv.org/abs/2609.27142

ROCS paper, code and dataset

• Paper https://arxiv.org/abs/2609.27142
• Code https://github.com/aalquwayfili/MINER
• Dataset https://huggingface.co/datasets/aalquwayfili/ROCS

Which sources cite ROCS?

• ROCS

Related General AI, Information retrieval, Vision & 3D, cs.CV benchmarks

• $\tau^\tau$-Bench: An Environment for End-To-End, Realistic Agent Construction
• 2d-fet-bench
• 360CityArena
• 3DCodeBench
• 5-Dialects-BN
• A Real-World CAD Segmentation Benchmark
• A3S-Bench
• AARRI-Bench