This paper investigates implicit causal graph construction from text by inferring intermediate causal events using Large Language Models (LLMs). It compares end-to-end graph construction with causal chain discovery methods and evaluates the validity of inferred causal relations against a manually curated database.
This paper explores detection within the stochastic block model with multiple clusters, providing proofs for achievability conjectures. It also discusses acyclic Belief Propagation and information-theoretic aspects of the model.
This content presents new approximation algorithms for fundamental graph problems, achieving improved performance. Specifically, it introduces a 7/5-approximation for graphic TSP, 3/2 for its path version, and 4/3 for the two-edge-connected subgraph problem.
Este artigo apresenta um levantamento e uma taxonomia abrangente de métodos de amostragem de grafos. O estudo organiza e classifica as diversas técnicas existentes, oferecendo uma visão estruturada sobre o tema.
This paper extends Courcelle's theorem by showing that models of MSO2 formulas with free variables can be represented with decision diagrams whose size is parameterized linearly. It establishes parameterized linear upper bounds for sentential decision diagrams (SDD) based on treewidth and ordered binary decision diagrams (OBDD) based on pathwidth.
AdaGraph is a graph-native clustering algorithm from the Structure-Centric Machine Learning (SC-ML) paradigm, which fundamentally dissolves the curse of dimensionality by replacing geometry-centric computation with topology-based computation. Operating within kNN graph topology, it requires no a priori specification of cluster numbers, handles noise, and scales effectively.
We propose PathBoost, a gradient tree boosting method for graph-level classification and regression, which learns discriminative path-based features directly from the input graph structure. This method introduces adaptations for binary classification, incorporates multiple node and edge attributes, and automatically selects anchor nodes, outperforming or matching graph neural networks and graph kernel approaches on several benchmark datasets.
This content provides a comprehensive survey of various algorithms designed to find the shortest path in graphs. It explores the methodologies, complexities, and practical applications of each algorithm, offering an in-depth view for students and researchers.
This document explores graph-theoretic models for predicting molecular measurements. The research focuses on applying mathematical structures to understand and anticipate chemical and biological properties.