AI Paper Atlas

Research map / Graph learning / Knowledge and temporal graphs

SQALER: Scaling Question Answering by Decoupling Multi-Hop and Logical Reasoning

Mattia Atzeni, Jasmina Bogojeska, Andreas Loukas

NeurIPS 2021 · Graph learning › Knowledge and temporal graphs · approach: Logical query answering and rule reasoning · 0 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 key-value memory network edge-level refinement to unrefined relation-level model (ablation)SQALER Unrefined · WebQSPHits@1: 70.6 -> 72.1Improved
add GNN edge-level refinement to unrefined relation-level model (ablation)SQALER Unrefined · WebQSPHits@1: 70.6 -> 76.1Improved
use GNN instead of KV-MemNN for refinement (ablation)SQALER · WebQSPHits@1: 72.1 -> 76.1Improved
add refinement step (KV-MemNN or GNN) on MetaQA (ablation)SQALER Unrefined · MetaQA 2 and 3Hits@1: 99.9 -> 99.9No clear effect
use relation-level model only versus EmQL on WebQSPSQALER Unrefined · WebQSPHits@1: 75.5 -> 70.6Hurt
use SQALER Unrefined instead of EmQL on MetaQA 3EmQL · MetaQA 3Hits@1: 99.1 -> 99.9Improved
use SQALER-GNN instead of PullNet on incomplete KG (50% edges)PullNet · WebQSP 50% KGHits@1: 50.3 -> 53.5Improved
use SQALER-GNN instead of PullNet on incomplete KG plus textPullNet · WebQSP 50% KG + TextHits@1: 51.9 -> 55.2Improved
add Wikipedia text documents to 50% KGSQALER -- GNN · WebQSPHits@1: 53.5 -> 55.2Improved
use SQALER-GNN instead of HPD on CFQ MCD-meanHPD · CFQaccuracy: 0.673 -> 0.671No clear effect
use SQALER-GNN instead of T5-11B-mod on CFQ MCD-meanT5-11B-mod · CFQaccuracy: 0.421 -> 0.671Improved
increase number of sampled relation sequences from 3 to 10 (ablation)SQALER relation-level model · WebQSPrecall: 0.91 -> 0.95Improved

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