AI Paper Atlas

Research map / Vision-language and multimodal / Visual question answering

Chain of Reasoning for Visual Question Answering

Chenfei Wu, Jinlai Liu, Xiaojie Wang, Xuan Dong

NeurIPS 2018 · Vision-language and multimodal › Visual question answering · approach: Universal VQA models · 39 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 chain structure (CoR-3) instead of stacked attention (Stack-3) (ablation)Mutan attention · VQA 2.0 valaccuracy: 63.90 -> 65.14Improved
use chain structure instead of parallel attention (Parallel-3) (ablation)Mutan attention · VQA 2.0 valaccuracy: 63.80 -> 65.14Improved
use chain structure with MLB attention (CoR-3) instead of plain MLB (ablation)MLB attention · VQA 2.0 valaccuracy: 62.91 -> 64.96Improved
use element-wise sum of guided objects for relations instead of RN-style concatenation (ablation)CoR-2 · VQA 2.0 valaccuracy: 62.46 -> 64.96Improved
add separate question guidances Gl and Gr for compound and initial objects (ablation)CoR-2 · VQA 2.0 valaccuracy: 64.73 -> 64.96Improved
include initial objects at each step's relational reasoning (ablation)CoR-2 · VQA 2.0 valaccuracy: 64.24 -> 64.96Improved
refine objects by weighting over previous compound objects (R_ij) rather than R_ji (ablation)CoR-2 · VQA 2.0 valaccuracy: 64.42 -> 64.96Improved
increase chain length from 2 to 3 sub-chains (ablation)CoR · VQA 2.0 valaccuracy: 64.96 -> 65.14Improved
use CoR-3 versus prior state of the artCoR-3 · TDIUCoverall accuracy: 85.03 -> 86.91Improved
use CoR-3 versus prior state of the artCoR-3 · COCO-QAaccuracy: 66.49 -> 69.38Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

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