Saved in:
Bibliographic Details
Main Authors: Ma, Zheng, Wang, Changxin, Ouyang, Yawen, Zhao, Fei, Zhang, Jianbing, Huang, Shujian, Chen, Jiajun
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2402.11572
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866929247736561664
author Ma, Zheng
Wang, Changxin
Ouyang, Yawen
Zhao, Fei
Zhang, Jianbing
Huang, Shujian
Chen, Jiajun
author_facet Ma, Zheng
Wang, Changxin
Ouyang, Yawen
Zhao, Fei
Zhang, Jianbing
Huang, Shujian
Chen, Jiajun
contents Evaluating the compatibility between textual descriptions and corresponding images represents a core endeavor within multi-modal research. In recent years, a proliferation of reference-free methods, leveraging visual-language pre-trained models (VLMs), has emerged. Empirical evidence has substantiated that these innovative approaches exhibit a higher correlation with human judgment, marking a significant advancement in the field. However, does a higher correlation with human evaluations alone sufficiently denote the complete of a metric? In response to this question, in this paper, we study if there are any deficiencies in reference-free metrics. Specifically, inspired by the Cobra Effect, we utilize metric scores as rewards to direct the captioning model toward generating descriptions that closely align with the metric's criteria. If a certain metric has flaws, it will be exploited by the model and reflected in the generated sentences. Our findings reveal that descriptions guided by these metrics contain significant flaws, e.g. incoherent statements and excessive repetition. Subsequently, we propose a novel method termed Self-Improving to rectify the identified shortcomings within these metrics. We employ GPT-4V as an evaluative tool to assess generated sentences and the result reveals that our approach achieves state-of-the-art (SOTA) performance. In addition, we also introduce a challenging evaluation benchmark called Flaws Caption to evaluate reference-free image captioning metrics comprehensively. Our code is available at https://github.com/aaronma2020/robust_captioning_metric
format Preprint
id arxiv_https___arxiv_org_abs_2402_11572
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Cobra Effect in Reference-Free Image Captioning Metrics
Ma, Zheng
Wang, Changxin
Ouyang, Yawen
Zhao, Fei
Zhang, Jianbing
Huang, Shujian
Chen, Jiajun
Computation and Language
Evaluating the compatibility between textual descriptions and corresponding images represents a core endeavor within multi-modal research. In recent years, a proliferation of reference-free methods, leveraging visual-language pre-trained models (VLMs), has emerged. Empirical evidence has substantiated that these innovative approaches exhibit a higher correlation with human judgment, marking a significant advancement in the field. However, does a higher correlation with human evaluations alone sufficiently denote the complete of a metric? In response to this question, in this paper, we study if there are any deficiencies in reference-free metrics. Specifically, inspired by the Cobra Effect, we utilize metric scores as rewards to direct the captioning model toward generating descriptions that closely align with the metric's criteria. If a certain metric has flaws, it will be exploited by the model and reflected in the generated sentences. Our findings reveal that descriptions guided by these metrics contain significant flaws, e.g. incoherent statements and excessive repetition. Subsequently, we propose a novel method termed Self-Improving to rectify the identified shortcomings within these metrics. We employ GPT-4V as an evaluative tool to assess generated sentences and the result reveals that our approach achieves state-of-the-art (SOTA) performance. In addition, we also introduce a challenging evaluation benchmark called Flaws Caption to evaluate reference-free image captioning metrics comprehensively. Our code is available at https://github.com/aaronma2020/robust_captioning_metric
title Cobra Effect in Reference-Free Image Captioning Metrics
topic Computation and Language
url https://arxiv.org/abs/2402.11572