Edge-efficient Transformer for real-time RF spectrum monitoring and modulation recognition.

A multi-agent framework for systematic repair strategy exploration in software agents.

A training-free retrieval framework for accurate skill retrieval in LLM agent libraries.

An open-source interactive long-horizon video world model based on a 15B video diffusion transformer.

A SPARQL extension for querying uncertain numeric measurement data in knowledge graphs.

A state-of-the-art open-weight model used for forensic analysis of AI-driven cyberattacks.

LLM-based framework for co-evolving placement objectives with routing feedback in chip design.

Integration of lookahead branching into neural network verifiers to speed up verification time.

An evolution-aware benchmark for GUI state-transition understanding for vision-language models.

An open-source vector index for cost-efficient and private retrieval in enterprise RAG systems.

Code for evaluating scientific memory in LLM agents.

A grounded code-agent platform for constructing editable LLM data pipelines

Framework for inducing evidence-grounded failure taxonomies from AI agent execution traces.

A clarification-aware and evolutionary 3D agent for resolving intent asymmetry in tool orchestration.

A data-driven game engine built in Rust

Self-hosted registry and analytics for AI Agents.

Alignment-Aware Bridge Matching for robust image-to-image translation with weakly aligned data.

A CIFAR-10-aligned synthetic image dataset for benchmarking AI-generated image detection.

Adaptive framework for multimodal survival analysis using Mamba architecture.

A dynamic benchmark for language models and deep-research agents on football forecasting.