Research paper on representation sensitivity in LLM-based mathematical problem solving.

The Open University Learning Analytics Dataset for dropout risk modelling.

A differentiable penalty that treats additivity as a measurable behavior of dense neural networks.

A hybrid framework integrating DQAOA with GPT-based quantum circuit generation.

An AI-powered platform for named-entity recognition in the crime domain.

A lightweight guard model for safe LLM agent tool calls.

An agentic dual-path framework designed to unmask visual deception in charts.

A statistics dashboard for Jellyfin logs

An open-source, local-first AI desktop client.

A multi-hop historical QA corpus for evaluating retrieval-augmented LLM systems in domain-specific contexts.

A generative model for global sea state estimation using diffusion processes.

Code and resources for studying prompt injection in automated résumé screening systems.

Offline reinforcement learning framework for optimizing warehouse SLAM throughput.

A Google Keep alternative using Glass design, built with React and Tailwind.

A self-hosted fitness tracker that integrates with various health data sources.

A free, open-source alternative to Crunchbase for startup data.

A theoretical framework for treating AI affect as a coordination layer in agentic collaboration.

Research on optimization dynamics and sparsity boundaries in large-scale pedestrian attribute recognition.

A benchmark specifically designed to evaluate cross-system urban requests and tool coverage.

A tool for easy gamma adjustments on KDE Plasma with Wayland.