Virtual Knapping (and Refitting) with Neural Networks: Proofs of Concept

Utilizing Concept Ontologies for Designing Neural Network-Based Classifiers

Kamil Szwed

Deep neural networks have achieved substantial success in image, text, and signal analysis, but their advantage is less consistent for heterogeneous tabular data, where tree-based ensemble methods often remain strong baselines. This study proposes CANON (Cross-Attention Neuro-sym…Read more

Unsupervised Deep-learning Methods for Low-dose Computed Tomography Reconstruction

Ran An

Computed tomography (CT) has become an indispensable imaging technique in medical diagnostics and industrial applications, owing to its non-invasive nature and high resolution in visualizing object internal structures. While X-ray CT (X-ray computed tomography) significantly enha…Read more

Also available via: Open MIND

Friction-Guided Inference: Calibrating Correction Strategies and Abstention from Logprob Signals

Tomas Pødenphant Lund

Large language models frequently possess the knowledge needed to answer a question correctly yet commit to the wrong response. This paper presents friction-guided inference, a calibrated inference-time pipeline that uses the model's own logprob distribution — available at zero co…Read more

From Sparse to Dense: Label-Efficient Weakly Supervised Segmentation for Images and Videos

J. Wang

Obtaining high-quality annotated data has become a primary bottleneck for training deep learning models, particularly for dense prediction tasks like semantic segmentation and video salient object segmentation. The demand for meticulous, pixel-level labeling makes fully-supervise…Read more

Also available via: Open MIND

SOAR: Smooth Online Activation Routing for Stable Neural Learning from Evolving Streams

Sizhen Niu

Online neural learning requires models that update after each incoming example, remain calibrated under distributional change, and avoid brittle gradient transmission. The original version of this work used a small static benchmark, a shallow model, few random seeds, and no signi…Read more

Also available via: Open MIND

HomoeoGWAS: subgenome-aware scalable LMM with DL prior conditional lift for allopolyploid GWAS

Shipeng Yang

HomoeoGWAS is a GWAS framework for allopolyploid crops (wheat AABBDD, cotton AADD, rapeseed AACC, oat AACCDD, strawberry octoploid). It combines subgenome-partitioned random-effect LMMs with per-logical-chromosome LOCO, an optional scope-conditional homoeolog Hadamard kernel, and…Read more