AI Paper Atlas

Research map / Large language models / Retrieval and knowledge

Explainable Neural Computation via Stack Neural Module Networks

Ronghang Hu, Jacob Andreas, Trevor Darrell, Kate Saenko

ECCV 2018 · Large language models › Retrieval and knowledge · approach: Knowledge graph reasoning with LLMs · 173 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 expert layout supervision (ablation)Stack-NMN · CLEVR val (VQA)accuracy (%): 93.0 -> 96.6Improved
add expert layout supervision (ablation)Stack-NMN · CLEVR-Ref val (REF)accuracy (%): 93.4 -> 96.0Improved
add joint VQA+REF training (no expert layout) (ablation)Stack-NMN · CLEVR val (VQA)accuracy (%): 93.0 -> 93.9Improved
add joint VQA+REF training (no expert layout) (ablation)Stack-NMN · CLEVR-Ref val (REF)accuracy (%): 93.4 -> 95.4Improved
add joint VQA+REF training (with expert layout) (ablation)Stack-NMN · CLEVR val (VQA)accuracy (%): 96.6 -> 96.5No clear effect
use soft differentiable layout instead of discrete layout, no expert layoutStack-NMN vs N2NMN · CLEVRaccuracy (%): 69.0 -> 93.0Improved
discretize module weights to one-hot at test time (ablation)Stack-NMN (no expert layout) · CLEVR val (VQA)accuracy (%): 93.0 -> 90.0Hurt
discretize module weights to one-hot at test time (ablation)Stack-NMN (expert layout) · CLEVR val (VQA)accuracy (%): 96.5 -> 94.8Hurt
add expert layout supervision (ablation)Stack-NMN · human forward prediction (failure detection)human accuracy: 0.625 -> 0.545Hurt
add expert layout supervision (ablation)Stack-NMN · human subjective understandingmean Likert score: 3.33 -> 3.47Improved
use modular Stack-NMN instead of non-modular MACinterpretability · human subjective understandingmean Likert score: 2.46 -> 3.33Improved
use Stack-NMN instead of MACVQA model · CLEVRaccuracy (%): 98.9 -> 93.0Hurt

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