graph neural network
PulseAugur coverage of graph neural network — every cluster mentioning graph neural network across labs, papers, and developer communities, ranked by signal.
- instance of alphaXiv 90%
- instance of machine learning 90%
- instance of graph attention network 90%
- used by knowledge graph 90%
- used by alphaXiv 70%
- instance of ScienceCast 70%
- used by ScienceCast 70%
- developed Gotit.pub 70%
- instance of Gotit.pub 70%
- used by CatalyzeX 70%
- instance of CatalyzeX 70%
- uses large-language models 70%
16 day(s) with sentiment data
-
New framework uses LLM and GNN for natural language crowd simulation control
Researchers have developed CrowdDirector, a novel framework designed to orchestrate crowd simulations in real-time using natural language directives. This system employs a large language model (LLM) to interpret user in…
-
Graph-based RL framework enhances power grid control
Researchers have developed a novel graph-based distributed reinforcement learning framework for managing modern power grids, addressing challenges posed by renewable energy integration and network expansion. This system…
-
New SymFP method uses spectral analysis to fingerprint symmetry in equivariant representations
Researchers have introduced Spectral Symmetry Fingerprints (SymFP), a novel method for characterizing symmetry in physical systems using equivariant representations. This technique leverages the singular-value spectra o…
-
Graph Neural Network Explanations Ambiguous Due to 'Deletion Fidelity' Definition
Researchers have identified a critical ambiguity in evaluating graph neural network (GNN) explanations, specifically concerning the concept of 'deletion fidelity.' This metric typically involves removing top-ranked edge…
-
Graph Neural Networks Compared to Chained-Join Features for Payroll Fraud Detection
A new research paper published on arXiv details a comparative study of methods for detecting payroll fraud within synthetic HR/ERP graphs. The study evaluated traditional analyst rules, explicit path features, and graph…
-
CIRCE system predicts LLM task complexity using constraint structure
Researchers have developed CIRCE, a novel system designed to predict the complexity of tasks for large language models (LLMs) before they attempt them. CIRCE analyzes the inherent constraint structure within a problem, …
-
New B-EAGLe framework enhances graph learning with sequential evidence accumulation
Researchers have introduced B-EAGLe, a novel probabilistic framework designed for graph-based learning in dynamic data-driven application systems. This framework reformulates graph classification as a sequential evidenc…
-
EnerGNN framework learns optimization-compatible energy functions for exact inference
Researchers have developed EnerGNN, a novel framework that learns energy functions for combinatorial optimization problems. This graph neural network-based approach integrates with mixed-integer programming to ensure ex…
-
Federated learning enhances cross-bank anti-money laundering detection
Researchers have developed a novel approach to Anti-Money Laundering (AML) using federated learning and graph neural networks. This method allows banks to detect money laundering patterns that span across separate trans…
-
New EH-Fusion framework tackles Euclidean crowding in hyperbolic graph learning
Researchers have developed a new framework called EH-Fusion to address Euclidean crowding in hyperbolic graph representation learning. This framework provides a geometric quantification of Euclidean crowding and a theor…
-
New method screens fairness edits in two-sided matching markets
Researchers have developed a closed-form linear projector that can reduce the influence of protected attributes in two-sided matching systems at inference time. This method aims to ensure fairness by bounding the welfar…
-
New Diffusion Reinforcement Learning Method Enhances 3D Bin Packing
Researchers have developed a new diffusion reinforcement learning method for online 3D bin packing problems. This approach integrates spatial heuristic strategies with a diffusion reinforcement learning framework to imp…
-
Graph Neural Networks Enhanced for Hyperbolic Transport Modeling with CFL Conditioning
Researchers have developed a new method for graph neural networks (GNNs) to better model hyperbolic partial differential equations (PDEs), which often involve complex phenomena like shocks and discontinuities. The propo…
-
Research paper reveals "token collapse" in graph soft prompting methods
A new research paper published on arXiv details a phenomenon called "token collapse" observed in textual graph soft prompting methods, specifically within the G-Retriever system. This collapse causes graph-specific toke…
-
New GNN-based method enhances handwritten music score synthesis for OMR
Researchers have developed a data-driven approach to improve the synthesis of handwritten music scores for Optical Music Recognition (OMR) training. By analyzing inter-glyph distributions from large datasets like MUSCIM…
-
New framework boosts financial crime detection using time-series models
Researchers have developed a new framework called Temporal-Context Edge Enrichment (TCEE) to improve financial crime detection. TCEE utilizes time-series foundation models (TSFMs) to summarize account transaction histor…
-
Hugging Face paper reveals "token collapse" in LLM graph prompting
A new paper from Hugging Face Daily Papers investigates the phenomenon of "token collapse" in graph soft prompting for large language models. This issue occurs when learned graph tokens become indistinguishable, functio…
-
New GNN Training Method Leverages Event Rate as a Feature
Researchers have developed a new method for training event-based Graph Neural Networks (GNNs) that leverages the event rate as a feature rather than treating it as a nuisance. This approach, called Curriculum FiLM Train…
-
New framework stabilizes graph filters for improved signal generation
Researchers have developed a new framework for designing graph filters to improve the generation of graph signals, particularly when the underlying graph structure is noisy or imperfectly estimated. The proposed method …
-
New graph neural network models interactions in complex chemical systems
Researchers have developed a new graph neural network designed to understand interactions within complex chemical systems composed of diverse substructures. This method addresses limitations in existing approaches that …