GPU-parallel robust control solver for nonlinear and NN dynamics with certified error bounds.

Logit-Contribution Scoring for identifying non-literal retrieval heads in LLMs

An RSS reader optimized for e-ink devices.

Competitive optimization of formally verified circuits

Large-scale benchmark for LLVM compiler issue resolution containing 423 validated tasks.

A columnar log storage and query engine designed for high efficiency and scalability.

Cloud implementation of the Model Context Protocol (MCP) for AI agents.

A black-box attack framework for jailbreaking function-calling LLMs via simulated moderation traces.

Agentic framework for iterative video retrieval and reasoning via soft query refinement.

Official code and videos for the Generalizable Data-efficient Agent (GenDa) framework.

A diagnostic egocentric video benchmark for evaluating embodied VLMs as runtime safety guards.

A self-governing context management system for reducing token usage and maintaining agent state in long-horizon tasks.

An automated agentic pipeline for generating verifiable, deterministic reaction rules for chemistry synthesis.

Medical Reasoning-aware Policy Optimization for multimodal LLMs in clinical image reasoning.

Framework for histogram-constrained image generation in diffusion models.

Uncertainty-guided synthetic context augmentation for semantic segmentation.

Zero-shot Entropy-Regularized Prompt Learning for Audio-Language Models.

A benchmark for measuring longitudinal psychometric stability (Mandate Salience Decay) in financial LLM agents.

PostgreSQL extensions managed as stackable Docker layers.

Robust statistical AI-generated text detection framework.