AI Paper Atlas

Research map / Vision-language and multimodal / Human and embodied modeling

MAGIC: Multimodal relAtional Graph adversarIal inferenCe for Diverse and Unpaired Text-Based Image Captioning

Wenqiao Zhang, Haochen Shi, Jiannan Guo, Shengyu Zhang, Qingpeng Cai, Juncheng Li, Sihui Luo, Yueting Zhuang

AAAI 2022 · Vision-language and multimodal › Human and embodied modeling · approach: Neural image captioning · 24 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
use MAGIC instead of ISM-LSTMText-based image captioning · TextCaps validationCIDEr: 31.1 -> 76.6Improved
add a central object-aware pool (ablation)MAGIC · TextCaps validationCIDEr: 71.3 -> 76.6Improved
add query attention over text tokens (ablation)MAGIC · TextCaps validationCIDEr: 75.9 -> 76.6Improved
use CGAN rather than the original GAN for alignment (ablation)MAGIC · TextCaps validationCIDEr: 65.3 -> 76.6Improved
add the language discriminator (ablation)MAGIC · TextCaps validationCIDEr: 73.9 -> 76.6Improved
vary COP capacity from N_k=3 to N_k=1 (ablation)MAGIC · TextCaps validationCIDEr: 76.6 -> 77.3Mixed
vary COP capacity from N_k=3 to N_k=2 (ablation)MAGIC · TextCaps validationCIDEr: 76.6 -> 75.8No clear effect
vary COP capacity from N_k=3 to N_k=4 (ablation)MAGIC · TextCaps validationCIDEr: 76.6 -> 75.3No clear effect
vary COP capacity from N_k=3 to N_k=5 (ablation)MAGIC · TextCaps validationCIDEr: 76.6 -> 73.4No clear effect
select central objects by learning their scores rather than using center, large-object, or random rules (ablation)MAGIC · TextCaps validationCIDEr: 72.3, 75.9, 67.8 -> 76.6Improved

Later papers that compared with this method

8 later accepted papers report MAGIC as a baseline in their results tables; 6 of them report a better number on most of the results they share with it.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy