Ziwei Zhao

Information density imbalance, not just instance count, is a key driver of category bias in object detection models, revealing new avenues for enhancing fairness and accuracy.

Protecting proprietary AI skills while enabling seamless multi-agent collaboration could redefine how we approach software engineering with AI.

Achieving up to 2x faster decoding in LLMs without additional training, Progressive Tree Drafting redefines how we approach speculative decoding.

Code agents are only eliminating 50% of code smells, revealing critical gaps in their understanding of maintainability and cross-file dependencies.