Ref-Adv: Exploring MLLM Visual Reasoning in Referring Expression Tasks

TL;DR

Ref-Adv introduces a challenging new benchmark for referring expression comprehension that emphasizes complex reasoning and grounding, revealing current models' reliance on shortcuts and highlighting areas for future improvement.

Contribution

The paper presents Ref-Adv, a novel REC benchmark designed to challenge models with nontrivial expressions and hard distractors, promoting genuine visual reasoning and grounding.

Findings

Models perform well on existing benchmarks but struggle on Ref-Adv.

Ref-Adv exposes reliance on shortcuts in current models.

Comprehensive analysis reveals gaps in visual reasoning capabilities.

Abstract

Referring Expression Comprehension (REC) links language to region level visual perception. Standard benchmarks (RefCOCO, RefCOCO+, RefCOCOg) have progressed rapidly with multimodal LLMs but remain weak tests of visual reasoning and grounding: (i) many expressions are very short, leaving little reasoning demand; (ii) images often contain few distractors, making the target easy to find; and (iii) redundant descriptors enable shortcut solutions that bypass genuine text understanding and visual reasoning. We introduce Ref-Adv, a modern REC benchmark that suppresses shortcuts by pairing linguistically nontrivial expressions with only the information necessary to uniquely identify the target. The dataset contains referring expressions on real images, curated with hard distractors and annotated with reasoning facets including negation. We conduct comprehensive ablations (word order perturbations and descriptor deletion sufficiency) to show that solving Ref-Adv requires reasoning beyond simple cues, and we evaluate a broad suite of contemporary multimodal LLMs on Ref-Adv. Despite strong results on RefCOCO, RefCOCO+, and RefCOCOg, models drop markedly on Ref-Adv, revealing reliance on shortcuts and gaps in visual reasoning and grounding. We provide an in depth failure analysis and aim for Ref-Adv to guide future work on visual reasoning and grounding in MLLMs.