A workflow for physics-audited agentic discovery in scientific machine learning.

Unified cross-paradigm benchmark for deepfake image detection evaluation.

A 3D dense captioning system combining PseudoCap data augmentation and VoxelCapNet.

Unsupervised anomaly detection of information operations users via behavioral and language patterns.

Research analyzing LLM conformity and the confound of repeated wrong answers.

A profiling tool for backend-specific energy attribution in mobile LLM inference.

A benchmark for risk identification and refusal behavior for biological research tasks.

Runtime verification framework for governed AI agent actions ensuring authorization and integrity.

An open-source ecosystem of plugins encoding scientific Python practice and research workflows into AI agents.

An open-source, tamper-evident runtime evidence framework for AI agents.

Clinical Decision Support System integrating deterministic analytics and RAG.

Research on research-trajectory hijacking attacks on deep research agents.

A Retrieval-Augmented Generation-based Examination Simulator acting as a high-fidelity clinical oracle.

A next-generation, Git-compatible version control system focusing on operation logs and easier revisions.

An experimental concatenative operating system.

A FOSS offline mapping solution.

A Jellyfin & Subsonic client for the terminal

A modular control panel for Linux server administration via Telegram.

An email alias service to protect privacy and manage professional email identities.

A benchmarking tool to score the performance of AI coding agents.