An open-source ecosystem of software components for developing smart city solutions and digital twins.

Trajectory-Relayed On-Policy Distillation for improving mathematical reasoning in LLMs.

A framework for using game solvers as turn-level teachers for LLM agent training.

Framework for training reasoning rerankers for tabular retrieval

A learned PSE encoder using polynomial filter banks for improved graph structural discrimination.

An auditable graph-evidence benchmark for phenotype-driven rare-disease diagnosis.

A framework for localization via audio-informed pooling in audio-visual retrieval models.

Agentic Vision-Language RAG for long, visually-rich multi-page documents.

A hierarchical skill graph framework for empowering LLM agents with reusable skill abstractions.

Dataset and platform for auditing pedagogical quality and risks in K-12 instructional explanations.

A black-box framework for inferring proprietary agent skills from execution trajectories.

Compute-aware evaluation framework for auditing Masked Diffusion Language Models remasking strategies.

A framework for scaling agentic models for full-scenario applications.

A practical reward framework for reinforcement learning in computer-use agents.

A single-block transformer for multi-entity reasoning focusing on structure over depth.

A decoding algorithm and SFT strategy for improving block diffusion LLMs.

A diffusion framework that embeds predictability asymmetry into trajectories for time series forecasting.

Study on LLM reliability beyond accuracy through meaning-preserving paraphrases

A diffusion-based foundation model for HPC telemetry and time-series forecasting.

OSS on-device voice dictation for macOS