graph neural networks

Manifesto of Symbiotic Homeoresis: Universal Civilizational Benchmark for Ethical AI, Salutogenesis, and Humanity Protection (Full Package & Axiomatic Core Digest)

Анотація Цей інтегрований пакет документів є єдиним суверенним депозитом, що містить повний текст та інваріантне категоріально-математичне ядро Маніфесту Симбіотичного Гомеорезу українською та англійською мовами. Праця проголошує фундаментальний гуманітарний та категоріальний зсув: Людина є єдиним суверенним Суб'єктом...

Feature-Refinement Classification Head (FCH) for Modeling Inter-Sample Relationships in Neural Networks

Conventional classification methods rely on the assumption that samples are independent and identically distributed, often ignoring the latent relational structures governing real-world data. Consequently, these methods fail to capture cross-sample dependencies, resulting in impoverished feature representations and sub...

Unbiased Recommender Systems with Implicit Feedback

Recommender systems typically rely on implicit feedback (e.g., clicks) to infer user preferences. However, such data is inherently prone to various biases, including position bias and popularity bias. Position bias occurs when higher-ranked items receive more interactions regardless of true relevance. Popularity bias r...

t2pmhc: A Structure-Informed Graph Neural Network to predict TCR-pMHC Binding

Packaging-only release — model code and bundled weights are identical to 1.1.2. Added pyproject.toml with complete PyPI metadata (dependencies, console entry point, bundled model data); replaces setup.py GitHub Actions workflow that publishes releases to PyPI via trusted publishing — t2pmhc is now installable with pip...

Graph neural networks versus feature-only learners for Chinese A-share sanction prediction: a holder–manager graph without transaction edges — reproducibility archive

Reproducibility companion to Graph neural networks versus feature-only learners for Chinese A-share sanction prediction: a holder–manager graph without transaction edges. Contains source, fixed-fit aggregates, source documentation, and tests for table regeneration. The reconstructed common-cutoff endpoint and graph sem...

marimo-flow

Reactive marimo notebooks for ML experimentation, with MLflow tracking, PINA physics-informed neural networks, and a multi-agent team built on pydantic-graph and Ollama Cloud.

Review of AI-driven discovery of cathode anode and electrolyte materials for advancing sodium-ion batteries

Sodium-ion batteries (SIBs) are emerging as a sustainable, cost-effective alternative to lithium-ion batteries (LIBs) for grid storage, electric vehicles, and electronics. However, commercialization depends on overcoming the performance bottlenecks in energy density, kinetic rates, and safety. Artificial Intelligence (...

Topological Proof Networks and Interdisciplinary Structural Isomorphisms: A Bourbaki 2.0 Framework for Categorical Graph Functor Closure and Discrete Metric Verification

Modern theoretical physics, pure mathematics, and artificial intelligence have converged upon a dual epistemological crisis: while automated neural theorem engines generate sprawling, opaque derivation steps that suffer from an unbridged "epistemic justification gap" (Tanswell & Berg, M\times\Phi 2026), human mathemati...

Community-Aware Dynamic Graph Learning for Brain Disorder Diagnosis

Trained checkpoints: graph neural network Hamiltonian and direct models for DNA charge transport

Trained PyTorch checkpoints for every run reported in the accompanying paper: graph neural network (GAT) models that predict DNA transmission and density-of-states spectra, either through a learned reduced-order Hamiltonian passed through an NEGF layer (Hamiltonian model) or directly from pooled graph features (direct...

SMART-KNEE: Sparse-Morphology Based Anatomical Reconstruction Using Topology-Aware Networks for Imageless Total Knee Arthroplasty

Structuring and Tokenizing Distributed User Interest Context for Generative Recommendation

Generative recommendation is an emerging paradigm that has shown promise in industrial recommendation systems, aiming to predict users’ next interactions from their historical behaviors. At the core of generative recommendation lies item tokenization, which bridges item semantics and recommendation models. However, exi...