Benchmark

RefCOCOg Benchmark

What is RefCOCOg?

RefCOCOg is a referring expression comprehension benchmark that evaluates spatial grounding in images. Given a natural language expression describing an…

Modality
multimodal
Evaluates
multimodal, grounding, vision
Openness
unknown
Importer
LLM Stats
Reported scores
1

Source provenance

RefCOCOg results and reported scores

LLM Stats

ModelOrganizationReported valueReportedEvidence
Nova 2 OmniAmazon0.8632025-12-02evidence

Scores are partitioned by the source that reported them and are never merged into a single cross-source ranking, because the sources measure different things and say so.

Which sources cite RefCOCOg?

Related multimodal, grounding, vision benchmarks