AI Paper Atlas

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

DecAF: Joint Decoding of Answers and Logical Forms for Question Answering over Knowledge Bases

Donghan Yu, Sheng Zhang, Patrick Ng, Henghui Zhu, Alexander Hanbo Li, Jun Wang, Yiqun Hu, William Yang Wang, Zhiguo Wang, Bing Xiang

ICLR 2023 · Vision-language and multimodal › Visual question answering · approach: Open-domain multimodal QA · 18 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
combine direct answers with answers from executed logical forms (ablation)DecAF · GrailQA devF1 overall: 54.7 -> 78.7; 72.4 -> 78.7Improved
use separate readers instead of one shared reader (ablation)DecAF · GrailQA devF1 overall: 78.7 -> 80.3Mixed
use separate readers instead of one shared reader (ablation)DecAF · CWQHits@1: 68.1 -> 66.5Hurt
increase the logical-form answer weight from 0.49 to 0.51 (ablation)DecAF answer combination · GrailQA devF1: 56.5 -> 78.7Improved
prefer the first executable logical-form answer, falling back to the first generated answer (ablation)DecAF answer combination · GrailQA devF1: 54.7 -> 78.7Improved
generate multiple direct answers instead of one (ablation)DecAF · GrailQA devF1 overall: 78.7 -> 78.1Hurt
increase the number of retrieved passages from 5 to 100 (ablation)DecAF · all four benchmarksQA performanceImproved
increase reader size from FiD-large to FiD-3B (ablation)DecAF · GrailQA testF1 overall: 76.0 -> 78.7Improved
use DPR instead of BM25 retrieval (ablation)DecAF · CWQHits@1: 68.1 -> 67.0Hurt
use DPR instead of BM25 retrieval (ablation)DecAF · WebQSPHits@1: 79.0 -> 80.7Improved
use DPR instead of BM25 retrieval (ablation)DecAF · FreebaseQAHits@1: 78.8 -> 79.0Improved
increase inference beam size from 1 to 20 (ablation)DecAF · GrailQA devF1Improved

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