UvA-DARE (Digital Academic Repository) EASTER: Learning to Split Transformers at the Edge Robustly

UvA-DARE (Digital Academic Repository) EASTER: Learning to Split Transformers at the Edge Robustly

Empowering edge intelligence through reparameterized lightweight transformers and distributed inference

A reparameterized transformer framework that integrates High-Rank Factorization (HRF) during training, layer merging at inference, and dynamic, load-balanced distributed inference across multiple devices is proposed, highlighting that reparameterized transformers, coupled with adaptive distributed inference and ultra-l...

EvEMTBench: An Open Benchmark for Machine Learning in Power System Protection

EvEMTBench makes generalization in machine-learning-based protection an explicit and reproducible evaluation problem and addresses an open, executable, and versioned benchmark that fixes these evaluation choices while leaving model design open.

Exposition on Over-Squashing Problem of GNNs: Current Methods, Benchmarks and Challenges

Graph-based message-passing neural networks (MPNNs) have achieved remarkable success in both node and graph-level tasks. However, several identified problems, including over-smoothing (OSM), limited expressive power, and over-squashing (OSQ), still restrain the performance of MPNNs. In particular, the latest identified...

HO-FL: Hybrid-Order Federated Learning for Heterogeneous Edge Devices

HO-FL is introduced, a hybrid-order FL framework that trains a model's bottom segment with ZO optimization and its top segment with FO optimization, and shows that hybrid-order local training can retain much of the full-FO performance with substantially lower client memory requirements.

When Guidance Goes Off-Scale: Recalibrating Diffusion Transformers under Analog Compute-in-Memory Nonidealities

The impact of analog CIM nonidealities on DiT sampling is characterized and a retraining-free, sampler-side recalibration that adjusts only the CFG scale for a given CIM condition is proposed, showing that the optimal guidance scale increases with CIM noise.