Vision-Language Models Are Not Pragmatically Competent in Referring Expression Generation

Ziqiao Ma, Jing Ding, Xuejun Zhang, Dezhi Luo, Jiahe Ding, Sihan Xu, Yuchen Huang, Run Peng, Joyce Chai

Keywords: Vision-Language Models, Pragmatics, Referring Expression Generation

TL;DR: We show significant pragmatic deficiencies in current VLMs when faced with referring expression generation compared to humans, as they violate Gricean maxims.

Abstract: Referring Expression Generation (REG) is a core task for evaluating the pragmatic competence of vision-language systems, requiring not only accurate semantic grounding but also adherence to principles of cooperative communication. However, current evaluations of vision-language models (VLMs) often overlook the pragmatic dimension, reducing REG to a region-based captioning task and neglecting Gricean maxims. In this work, we revisit REG from a pragmatic perspective, introducing a new dataset (RefOI) of 1.5k images annotated with both written and spoken referring expressions. Through a systematic evaluation of state-of-the-art VLMs, we identify three key failures of pragmatic competence: (1) failure to uniquely identify the referent, (2) inclusion of excessive or irrelevant information, and (3) misalignment with human pragmatic preference, such as the underuse of minimal spatial cues. We also show that standard automatic evaluations fail to capture these pragmatic violations, reinforcing superficial cues rather than genuine referential success. Our findings call for a renewed focus on pragmatically informed models and evaluation frameworks that align with real human communication.

Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the COLM Code of Ethics on https://colmweb.org/CoE.html

Author Guide: I certify that this submission complies with the submission instructions as described on https://colmweb.org/AuthorGuide.html

Submission Number: 1112