Experience

Huawei 2012 Theory Lab

Researcher

Microsoft Research & Ads

Research intern · Information retrieval for ads and language models

Research

01

Data

Investigating scaling laws and data-mixing laws for pre-training, as well as supervised fine-tuning, RLHF and RLVR, reasoning and distillation data, retrieval-augmented generation, and synthetic data for post-training.

02

Efficiency

Improving model efficiency through MoE expert placement and load balancing, alongside asynchronous NPU scheduling across computation, communication, and multi-stage execution.

03

Retrieval

Semantic identifiers and sequential transduction for generative retrieval, exploring scalable models for large-scale recommendation, search, and advertising.

Publications

Publications are being updated. For the latest work or collaboration enquiries, please get in touch.