AI Paper Atlas

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

Implicit Differentiable Outlier Detection Enable Robust Deep Multimodal Analysis

Zhu Wang, Sourav Medya, Sathya N. Ravi

NeurIPS 2023 · Vision-language and multimodal › Visual question answering · approach: Multimodal LLM question answering · 1 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 knowledge graph triplets (ablation)VK-OOD · VQAv2accuracy: 73.9 -> 74.6Improved
add OOD detection layer only (ablation)VK-OOD · VQAv2accuracy: 73.9 -> 74.1Improved
add both KG and OOD layer (ablation)VK-OOD · VQAv2accuracy: 73.9 -> 76.8Improved
add knowledge graph triplets (ablation)VK-OOD · OKVQAaccuracy: 45.5 -> 48.3Improved
add OOD detection layer only (ablation)VK-OOD · OKVQAaccuracy: 45.5 -> 46.2Improved
add both KG and OOD layer (ablation)VK-OOD · OKVQAaccuracy: 45.5 -> 52.4Improved
add both KG and OOD layer (ablation)VK-OOD · NLVR2accuracy: 80.6 -> 83.9Improved
increase number of retrieved knowledge tripletsVK-OOD · OKVQAaccuracy: 52.4Improved
add BERT knowledge encoder to ConceptNet (ablation)VK-OOD · OKVQAaccuracy: 51.1 -> 52.4Improved
30% incomplete knowledge tripletsVK-OOD · OKVQAaccuracy: 52.4 -> 51.2Hurt
add OOD layer to ITM-only pretraining (ablation)VK-OOD · Flickr30k zero-shotTR@1: 82.1 -> 84.5Improved
add OOD layer to ITM+MLM pretraining (ablation)VK-OOD · Flickr30k zero-shotIR@1: 74.1 -> 77.2Improved

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