AI Paper Atlas

Research map / Graph learning / Graph self-supervised learning

Beyond Flat Walks: Compositional Abstraction for Autoregressive Graph Generation

Kaiwen Bian, Andrew H Yang, Ali Parviz, Gal Mishne, Yusu Wang

NeurIPS 2026 · Graph learning › Graph self-supervised learning · approach: Diffusion-based graph generation · 0 citations (OpenAlex)

Read the paper

AI-extracted summary. Written by AI from the paper's full text. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Proposes a hierarchical, compositional abstraction for autoregressive graph generation that replaces flat node-edge walks, aimed at large structurally complex biological graphs.

Problem

Autoregressive graph generation linearizes graphs into flat walks over nodes and edges, which scales poorly to large, structurally complex biological graphs where higher-order organization is dense and important.

Approach

Introduces a hierarchical graph abstraction that composes higher-order structures so the autoregressive generation operates over abstracted components rather than a flat node-edge sequence.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
replace flat SENT walk with hierarchical HDT-HAC-RSENT · Five molecular benchmarksFCD: 51.6% to 64.6% lower FCDImproved
replace flat SENT walk with hierarchical HDT-HAC-RSENT · Five molecular benchmarksMoleculePGD: 34.3% to 86.6% lowerImproved
replace SENT with HDT-HAC-R (within-community randomization)SENT · Five molecular benchmarksV.U.N.: 3.6% to 37.2% deficitHurt
use hierarchy-wide randomization (FR) instead of within-community (R) (ablation)HDT-HAC · Five molecular benchmarksV.U.N.: Macrocycle 0.537 -> 0.762Improved
replace SENT with HDT-HAC-R for zero-shot motif promptingSENT · Motif-conditioned generationmotif discrepancy / V.U.N.Hurt
replace SENT with HDT-HAC-RSENT · Proteins (D&D)V.U.N.: +62.8%Improved
replace SENT with HDT-HAC-RSENT · SSBMV.U.N.: +59.3%Improved
replace SENT with default HDT-HAC-RSENT · SBMV.U.N.: 0.669 -> 0.276Hurt
supply planted hierarchy instead of HAC at tau=10 (FR) (ablation)HDT-HAC-FR · Connected RSBM (R=4)V.U.N.: 0.083 -> 0.744Improved
increase HAC stopping threshold (ablation)HDT-HAC · SBMV.U.N.Improved
use hierarchical serializationSENT · Planargraph MMDHurt
replace HDT with declarative HSENT or typed HDTC (ablation)HDT-R · GuacaMol and MOSESvalidity, V.U.N., MoleculePGDHurt

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