arXiv AI

Hierarchical GNNs for power flow: letting physics shape the hierarchy

The paper presents a hierarchical graph neural network (GNN) approach for power‑flow modeling that leverages physics‑informed graph reductions. By exchanging information through two reduced graphs within the corrective network of GENCO, the model achieves superior generalization across three grid topologies and new operating scenarios, outperforming both the flat baseline and a Quotient construction. Training required only about 200 epochs and fewer than 1,900 scenarios per grid, yet the hierarchical models achieved a macro family‑balanced voltage error of 0.851±0.110, a 51.3% improvement over the per‑bus mean fitted on training solutions.

arXiv Machine Learning

Sep 23

Towards Hierarchical GNNs for multi-grid power flow: generalization across operating scenarios

The paper proposes a hierarchical graph neural network (GNN) architecture for multi‑grid power‑flow modeling that exchanges information through two reduced graphs within a GENCO‑based corrective network. Experiments on three grid topologies show that the Kron‑derived transport approach reduces macro family‑balanced voltage error by 85 % compared to a flat backbone and 31 % compared to a Quotient construction, outperforming per‑bus mean baselines across all seeds. Preliminary results indicate strong generalization across operating scenarios within the studied topologies, though cross‑topology transfer remains an open challenge.

By Carmine Delle Femine, Leire Garin Atxaga, Asier Diaz-Iglesias, Juan Pablo Maroto Herrera, Ane Miren Florez-Tapia, Marco Quartulli. Izaro Goienetxea Urziku

arXiv AI

Jun 24

Decentralised AI Training and Inference with BlockTrain

arXiv:2606. 24722v1 Announce Type: new Abstract: Frontier AI training is increasingly shaped by access to dense, centrally controlled accelerator clusters.

By Peter Toth

llms

arXiv AI

Jul 20

Diagnosing Overhead in Dispatch Operations: Cross-architecture Observatory

arXiv:2605. 20982v2 Announce Type: replace-cross Abstract: AlltoAll dispatch is the dominant bottleneck of MoE expert parallelism, and the interconnect community has responded with four families of mitigations: predictive sample placement, adaptive expert relayout, hierarchical collectives, and EP-aware topology.

By Bole Ma, Jan Eitzinger, Harald Koestler, Gerhard Wellein

benchmarks

arXiv Machine Learning

2d ago

Scaling Zero-Order Pretraining through Model Sharding

arXiv:2609.37899v1 Announce Type: new Abstract: Zero-order optimization (ZO) trains without backpropagation, making it relevant to forward-only hardware and non-differentiable loss, but its gradient...

By Francois Chaubard, Mykel J. Kochenderfer, Chris R\'e

Hugging Face Trending Papers

Jul 15

MxGPS: Multiplex Graph Transformers for a Power Grid Foundation Model

Single-task fine-tuning of graph neural networks (GNNs) for power grid problems exhibits a systematic failure mode: models that achieve the lowest in-distribution error degrade the most under topology shift. We term this topology overfitting: the tendency of task-specific gradient signals to encode relational structure particular to the training topologies rather than the underlying physics, causing models to fail on unseen grids despite strong in-distribution performance.

llms fine-tuning benchmarks

arXiv AI

Jul 16

MxGPS: Multiplex Graph Transformers for a Power Grid Foundation Model

arXiv:2607. 13763v1 Announce Type: cross Abstract: Single-task fine-tuning of graph neural networks (GNNs) for power grid problems exhibits a systematic failure mode: models that achieve the lowest in-distribution error degrade the most under topology shift.

By Charilaos Papaioannou, Ioannis Tsantilas, Dimitris Giannakakos, Vasilis Michalakopoulos, Sotiris Pelekis, Vangelis Marinakis, Arsam Aryandoust, Antonello Monti, Ricardo J. Bessa, Perdo P. Vergara, Jochen Cremer, Elissaios Sarmas

llms fine-tuning benchmarks

arXiv Machine Learning

Jul 15

Gradient-Free Topology Adaptation for Power Flow Surrogates via In-Context Whitening

arXiv:2607. 12241v1 Announce Type: cross Abstract: Machine-learned surrogates for the AC power flow (ACPF) problem amortize the cost of repeated solves on a fixed network, but lose one to two orders of magnitude of accuracy when a line outage changes the topology.

By Ayushi Jolotia, Parikshit Pareek

arXiv AI

2d ago

Scalable Heterogeneous Graph Foundation Models for Data-Driven Optimal Power Flow in Smart Grids

arXiv:2605.23194v2 Announce Type: replace-cross Abstract: Fast and reliable optimal power flow (OPF) approximation is important for power system operation, yet heterogeneous OPF graph models are ofte...

By Massimiliano Lupo Pasini, Yijiang Li, Kibaek Kim, Teja Kuruganti

llms fine-tuning benchmarks

arXiv AI

Jun 3

Fine-Tuning and Serving Gemma 4 31B on Google Cloud TPU: A Technical Comparison with GPU Baselines

arXiv:2605. 25645v2 Announce Type: replace-cross Abstract: We present the first end-to-end demonstration of fine-tuning and serving Google's Gemma 4 31B model on TPU hardware, providing an empirical comparison of TPU and GPU platforms for large language model adaptation.

By Jatin Kishnani, Mayank Goel, Amit Singh, Pulkit Agrawal, Sairanjan Mishra

llms fine-tuning efficiency data-engineering

arXiv AI

Jul 24

When Does Recurrence Become an Algorithm? Convergence Selection in Weight-Tied Looped Transformers

arXiv:2607. 20594v1 Announce Type: cross Abstract: When does a weight-tied looped transformer -- one block applied T times -- implement an actual algorithm?

By Tong Zhang, Junhao Hu, Yun Peng, Tao Xie

llms benchmarks

arXiv Machine Learning

Jun 8

Reversible Foundations: Training a 120B Sparse MoE through State-Preserving Scaling

arXiv:2606. 07404v1 Announce Type: new Abstract: This paper reports on training a hundred-billion-parameter sparse mixture of experts on a single eight-GPU node, end to end.

By Rohan Shravan

llms nlp efficiency

arXiv AI

Jul 3

Do Newer Lightweight CNNs Perform Better Under Resource Constraints? A Controlled Multigenerational Study of Architecture, Initialization, Training Budget, and Efficiency

arXiv:2607. 01984v1 Announce Type: cross Abstract: Newer lightweight convolutional neural networks are often presented as improving predictive performance and deployment efficiency, but such claims require controlled evaluation.

By Tasnim Shahriar

Related stories