AI Paper Atlas

Research map / Video / Video-language understanding

Adaptive Hierarchical Graph Reasoning With Semantic Coherence for Video-and-Language Inference

Juncheng Li, Siliang Tang, Linchao Zhu, Haochen Shi, Xuanwen Huang, Fei Wu, Yi Yang, Yueting Zhuang

ICCV 2021 · Video › Video-language understanding · approach: Long video understanding · 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
add hierarchical graph network (AHGN) over MTS (ablation)MTS · VIOLIN testaccuracy: 67.60 -> 69.76Improved
add cross-modal OT coherence loss (ablation)AHGN · VIOLIN testaccuracy: 69.76 -> 70.19Improved
add cross-level mutual information loss (ablation)AHGN · VIOLIN testaccuracy: 69.76 -> 70.47Improved
add both coherence losses (ablation)AHGN · VIOLIN testaccuracy: 69.76 -> 71.38Improved
use adaptive instead of fixed graph structure (ablation)Backbone · VIOLIN testaccuracy: 64.65 -> 65.32Improved
use adaptive instead of fixed graph structure (ablation)AHGN · VIOLIN testaccuracy: 68.91 -> 69.76Improved
add gated inter-modal message passing (GER) (ablation)Backbone · VIOLIN testaccuracy: 65.32 -> 67.07Improved
add gated intra-modal message passing (GRA) (ablation)Backbone · VIOLIN testaccuracy: 65.32 -> 66.74Improved
add temporal-level reasoning (ablation)Backbone · VIOLIN testaccuracy: 65.32 -> 66.93Improved
add temporal-level reasoning to GER + GRA (ablation)Backbone + GER + GRA · VIOLIN testaccuracy: 68.03 -> 69.76Improved
replace GloVe with BERT text features (ablation)MTS · VIOLIN testaccuracy: 60.33 -> 67.60Improved
replace ResNet image features with detection features (ablation)MTS · VIOLIN testaccuracy: 67.60 -> 67.84No clear effect

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