AI Paper Atlas

Research map / Graph learning / GNN architectures

Unsat Core Prediction through Polarity-Aware Representation Learning over Clause-Literal Hypergraphs

Zhenchao Sun, Shuai Ma, Ping Lu, Chongyang Tao

ICML 2026 · Graph learning › GNN architectures · approach: Graph structure learning · 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
replace bipartite graph with hypergraph message passing (ablation)NeuroCore-based paSAT · CA unsat-coreTop-M precision (avg): 0.238 -> 0.360Improved
add invariant-equivariant variable decomposition (ablation)paSAT (HG) · CA unsat-coreTop-M precision (avg): 0.360 -> 0.383Improved
add polarity-inversion consistency regularization (ablation)paSAT (HG+DE) · CA unsat-coreTop-M precision (avg): 0.383 -> 0.408Improved
replace bipartite graph with hypergraph message passing (ablation)NeuroCore-based paSAT · SR unsat-coreTop-M precision (avg): 0.895 -> 0.898Improved
add decomposition and regularization to hypergraph model (ablation)paSAT (HG) · SR unsat-coreROC-AUC (avg): 0.808 -> 0.854 -> 0.872Improved
replace bipartite graph with hypergraph message passing (ablation)NeuroCore-based paSAT · PS unsat-coreTop-M precision (avg): 0.774 -> 0.784Improved
add decomposition and regularization to hypergraph model (ablation)paSAT (HG) · PS unsat-coreTop-M precision (avg): 0.784 -> 0.792 -> 0.797Improved
use paSAT instead of NeuroCoreunsat-core predictor · CA unsat-coreTop-M precision (avg): 0.238 -> 0.408Improved
use paSAT instead of NeuroCoreunsat-core predictor · SR unsat-corePR-AUC (avg): 0.938 -> 0.963Improved
use SATFormer instead of GCNunsat-core predictor · SR unsat-coreTop-M precision (avg): 0.880 -> 0.881No clear effect
guide Glucose with paSAT variable scoresGlucose · SATCOMP 2023instances solved: 133 -> 148Improved
guide Glucose with paSAT variable scoresGlucose · SATCOMP 2024instances solved: 111 -> 118Improved

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