Search:
Match:
5 results
Research Paper#Power Systems, Graph Neural Networks, Data Reconstruction🔬 ResearchAnalyzed: Jan 3, 2026 06:31

GNN with Auxiliary Learning for PMU Data Reconstruction

Published:Dec 31, 2025 01:00
1 min read
ArXiv

Analysis

This paper addresses the critical problem of missing data in wide-area measurement systems (WAMS) used in power grids. The proposed method, leveraging a Graph Neural Network (GNN) with auxiliary task learning (ATL), aims to improve the reconstruction of missing PMU data, overcoming limitations of existing methods such as inadaptability to concept drift, poor robustness under high missing rates, and reliance on full system observability. The use of a K-hop GNN and an auxiliary GNN to exploit low-rank properties of PMU data are key innovations. The paper's focus on robustness and self-adaptation is particularly important for real-world applications.
Reference

“The paper proposes an auxiliary task learning (ATL) method for reconstructing missing PMU data.”

PermalinkArXiv
Research Paper#AI, Venture Capital, Startup Prediction, Multi-Agent Systems🔬 ResearchAnalyzed: Jan 3, 2026 19:51

LLM Agents Predict Startup Success: A Collective Simulation Approach

Published:Dec 27, 2025 14:34
1 min read
ArXiv

Analysis

This paper addresses the critical challenge of predicting startup success, a high-stakes area with significant failure rates. It innovates by modeling venture capital (VC) decision-making as a multi-agent interaction process, moving beyond single-decision-maker models. The use of role-playing agents and a GNN-based interaction module to capture investor dynamics is a key contribution. The paper's focus on interpretability and multi-perspective reasoning, along with the substantial improvement in predictive accuracy (e.g., 25% relative improvement in precision@10), makes it a valuable contribution to the field.
Reference

“SimVC-CAS significantly improves predictive accuracy while providing interpretable, multiperspective reasoning, for example, approximately 25% relative improvement with respect to average precision@10.”

PermalinkArXiv
Research Paper#Molecular Dynamics, Graph Neural Networks, Surrogate Models🔬 ResearchAnalyzed: Jan 4, 2026 00:02

GNN Surrogate Models for Accelerated Molecular Dynamics

Published:Dec 26, 2025 01:22
1 min read
ArXiv

Analysis

This paper introduces a graph neural network (GNN) based surrogate model to accelerate molecular dynamics simulations. It bypasses the computationally expensive force calculations and numerical integration of traditional methods by directly predicting atomic displacements. The model's ability to maintain accuracy and preserve physical signatures, like radial distribution functions and mean squared displacement, is significant. This approach offers a promising and efficient alternative for atomistic simulations, particularly in metallic systems.
Reference

“The surrogate achieves sub angstrom level accuracy within the training horizon and exhibits stable behavior during short- to mid-horizon temporal extrapolation.”

PermalinkArXiv
Research#llm🔬 ResearchAnalyzed: Dec 25, 2025 02:19

A Novel Graph-Sequence Learning Model for Inductive Text Classification

Published:Dec 24, 2025 05:00
1 min read
ArXiv NLP

Analysis

This paper introduces TextGSL, a novel graph-sequence learning model designed to improve inductive text classification. The model addresses limitations in existing GNN-based approaches by incorporating diverse structural information between word pairs (co-occurrence, syntax, semantics) and integrating sequence information using Transformer layers. By constructing a text-level graph with multiple edge types and employing an adaptive message-passing paradigm, TextGSL aims to learn more discriminative text representations. The claim is that this approach allows for better handling of new words and relations compared to previous methods. The paper mentions comprehensive comparisons with strong baselines, suggesting empirical validation of the model's effectiveness. The focus on inductive learning is significant, as it addresses the challenge of generalizing to unseen data.
Reference

“we propose a Novel Graph-Sequence Learning Model for Inductive Text Classification (TextGSL) to address the previously mentioned issues.”

PermalinkArXiv NLP
Research#GNN🔬 ResearchAnalyzed: Jan 10, 2026 07:47

Advancing Aerodynamic Modeling with AI: A Multi-fidelity Dataset and GNN Surrogates

Published:Dec 24, 2025 04:53
1 min read
ArXiv

Analysis

This research explores the application of Graph Neural Networks (GNNs) for creating surrogate models of aerodynamic fields. The paper's contribution lies in the development of a novel dataset and empirical scaling laws, potentially accelerating design cycles.
Reference

“The research focuses on a 'Multi-fidelity Double-Delta Wing Dataset' and its application to GNN-based aerodynamic field surrogates.”

PermalinkArXiv