Physics-informed neural network for sub-surface mineral prospectivity modelling.

Omnimodal agent orchestration framework for multi-agent collaboration across text, image, audio, and video.

GitHub action to review pull requests for analytics instrumentation issues.

A PyTorch-native library for efficient video and audio decoding for AI training.

Detects AI/LLM contributions in git repositories via commit history analysis

Open-Source Silicon Initialization Library for low-level firmware validation.

An open-source autonomous agent framework capable of extending its own capabilities via package installation.

A structured generation language for LLMs that integrates the WitCert observability meter.

A PoC Dolby Atmos encoder

Knowledge Restoration-driven Prompt Optimization framework for open-domain relational triplet extraction.

A graph-based malware detection method utilizing operational role identification and representation learning.

Code and deterministic analysis pipeline for runtime data-quality gating in agentic AI.

Open-source automated auditing framework for evaluating deceptive and scheming behaviors in LLMs.

A framework for automating complex scientific workflows, such as atomistic simulations.

A self-hosted centralized hub for tracking physical and digital collections.

A protocol and deployable guard to measure and bound unsupported citations in agentic scientific synthesis.

A physician-annotated benchmark of LLM reasoning traces from real clinical cases.

A protocol for knowledge-centric self-improvement in agentic AI systems.

Multi-hypergraph boosted multi-interest self-supervised learning for conversational recommendation.

Extension of LeWorldModel adding high-level planning over latent subgoals for control.