aiqbench analysis
RefBench-PRO introduces a novel evaluation framework for Multi-modal Large Language Models that decomposes referring expression comprehension into distinct perception and reasoning dimensions. By automating data generation across six specific sub-tasks, this benchmark offers an interpretable scoring mechanism to better assess the grounding capabilities of MLLMs beyond simple perceptual checks. The inclusion of a reinforcement learning-based training scheme establishes a new baseline for improving localization accuracy under complex cognitive conditions.
RefBench-PRO: Perceptual and Reasoning Oriented Benchmark for Referring Expression Comprehension
arXiv:2512.06276v3 Announce Type: replace-cross Abstract: Referring Expression Comprehension (REC) is a vision-language task that localizes a specific image region based on a textual description. Existing REC benchmarks primarily evaluate perceptual capabilities and lack interpretable scoring mechanisms, which cannot reveal the grounding capability of Multi-modal Large Language Model (MLLM) across different cognitive abilities. To address this limitation, we introduce RefBench-PRO, a comprehensive REC benchmark, which decomposes referring expressions into two core dimensions, i.e., perception and reasoning, and further subdivides them into six progressively challenging tasks, such as attribute, position, interaction, commonsense, relation and reject. We also develop a fully automated data-generation pipeline that produces diverse referring expressions across these six sub-dimensions. Furthermore, We propose Ref-R1, an RL-based learning scheme, which incorporates Dynamic IoU-based GRPO to improve localization accuracy under increasingly complex reasoning conditions, establishing a stronger baseline for REC. Extensive experiments demonstrate that our RefBench-PRO enables interpretable evaluation of MLLM on referring expression comprehension, presenting greater challenges in both perception and reasoning.
Abstract:Referring Expression Comprehension (REC) is a vision-language task that localizes a specific image region based on a textual description. Existing REC benchmarks primarily evaluate perceptual capabilities and lack interpretable scoring mechanisms, which cannot reveal the grounding capability of Multi-modal Large Language Model (MLLM) across different cognitive abilities. To address this limitation, we introduce RefBench-PRO, a comprehensive REC benchmark, which decomposes referring expressions into two core dimensions, i.e., perception and reasoning, and further subdivides them into six progressively challenging tasks, such as attribute, position, interaction, commonsense, relation and reject. We also develop a fully automated data-generation pipeline that produces diverse referring expressions across these six sub-dimensions. Furthermore, We propose Ref-R1, an RL-based learning scheme, which incorporates Dynamic IoU-based GRPO to improve localization accuracy under increasingly complex reasoning conditions, establishing a stronger baseline for REC. Extensive experiments demonstrate that our RefBench-PRO enables interpretable evaluation of MLLM on referring expression comprehension, presenting greater challenges in both perception and reasoning.
Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
arXiv:2512.06276 [cs.CV]
(or arXiv:2512.06276v3 [cs.CV] for this version)
Full text under the publisher's open licence · 173 words
Related coverage
[AINews] Meta Muse Wonder & Wonder Full: MSL’s Claude 5.5-tier Frontier Model
Latent Space · Oct 13, 2026
State of the Field: AI for Epistemics and Coordination
LessWrong · Oct 2, 2026
LiveLaw Supreme Court Half-Yearly Digest: January - June 2026 [Part 2]
livelaw.in · Sep 13, 2026
Article: Your Next DSL Author is a Language Model
InfoQ · Sep 16, 2026