Heng Chang

AnchorSteer introduces a framework that achieves high-fidelity, structure-preserving music editing by decoupling semantic concept injection from structural constraints.

HarnessForge introduces a meta-adaptive framework that jointly evolves the execution structure (harness) and the reasoning policy of LLM agents, significantly improving overall system performance across diverse tasks.

This paper introduces a distribution-aware framework for modeling and benchmarking Mixture-of-Experts (MoE) inference, showing that the best fused-MoE kernel changes with routing skew and token count, and presents DA-MoE, a GPU-resident kernel-dispatch runtime that improves geomean fused-MoE latency.