AI Paper Atlas

Research map / Graph learning / Knowledge and temporal graphs

Generative Representation Learning on Hyper-relational Knowledge Graphs via Masked Discrete Diffusion

Jaejun Lee, Seheon Kim, Joyce Jiyoung Whang

ICML 2026 · Graph learning › Knowledge and temporal graphs · approach: Knowledge graph embeddings · 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 contextual message passing (context messages) (ablation)KREPE · WikiPeople- fact generation (scratch)accuracy: 0.578 -> 0.855Improved
add stochastic structure sampling (ablation)KREPE · WD50K entity prediction (all positions)MRR: 0.296 -> 0.419Improved
use attention instead of mean pooling (ablation)KREPE · WikiPeople- entity prediction (all positions)MRR: 0.466 -> 0.522Improved
use shared initialization instead of individual initialization (ablation)KREPE · WikiPeople- entity prediction (all positions)MRR: 0.390 -> 0.522Improved
replace AO-AR diffusion loss with link prediction cross-entropy (ablation)KREPE · WikiPeople- fact generation (scratch)accuracy: 0.855 -> 0.002Hurt
replace AO-AR diffusion loss with link prediction cross-entropy (ablation)KREPE · WD50K entity prediction (all positions)MRR: 0.419 -> 0.415No clear effect
use dot product instead of cosine similarity (ablation)KREPE · WikiPeople- entity prediction (all positions)MRR: 0.497 -> 0.522Improved
use KREPE instead of LLM baselines for fact generationfact generation · WikiPeople- fact generation (scratch)accuracy: 0.369 -> 0.855Improved
use KREPE instead of MAYPL for entity predictionlink prediction · WikiPeople- entity prediction (primary)MRR: 0.519 -> 0.519No clear effect
use KREPE instead of MAYPL for entity predictionlink prediction · WD50K entity prediction (primary)MRR: 0.381 -> 0.389Improved
use KREPE instead of HDiff for relation predictionlink prediction · WD50K relation prediction (primary)MRR: 0.950 -> 0.963Improved
use KREPE instead of best LLM baseline for arbitrary-masking generationfact generation · WD50K fact generation (arbitrary masking)accuracy: 0.604 -> 0.604No clear effect

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