About me

I am a Ph.D. student at HKBU, advised by Professor Bo Han and Professor Yang Liu. Previously, I worked or interned at Tencent YouTu Lab, collaborating with Fanxu Meng, Ke Li, Xing Sun; ByteDance AI Lab, collaborating with Xiaoying Zhang, Yang Liu; Alibaba DAMO Academy, collaborating with Qingsong Wen, Liang Sun.

I focus on Data-Centric AI, with the goal of developing robust and reliable data engines that drive AI systems. My work has been published in top-tier AI and CV venues, including ICLR, NeurIPS, ICML, and CVPR, with 10+ first author or co-first author papers and 2000+ citations.

Selected Publications

Task-Aware Data Selection via Proxy-Label Enhanced Distribution Matching for LLM Finetuning (to appear on ICLR 2026)
Hao Cheng*, Rui Zhang*, Ling Li, Na Di, Jiaheng Wei, Zhaowei Zhu, Bo Han.

RobustTSF: Towards Theory and Design of Robust Time Series Forecasting with Anomalies (ICLR 2024)
Hao Cheng, Qingsong Wen, Yang Liu, Liang Sun.
paper code

Identifiability of Label Noise Transition Matrix (ICML 2023)
Yang Liu, Hao Cheng, Kun Zhang.
paper code

Mitigating Memorization of Noisy Labels via Regularization between Representations (ICLR 2023)
Hao Cheng*, Zhaowei Zhu*, Xing Sun, Yang Liu.
paper code

Learning with Instance-Dependent Label Noise: A Sample Sieve Approach (ICLR 2021)
Hao Cheng*, Zhaowei Zhu*, Xingyu Li, Yifei Gong, Xing Sun, Yang Liu.
paper code

Pruning Filter in Filter (NeurlPS 2020)
Fanxu Meng*, Hao Cheng*, Ke Li, Huixiang Luo, Xiaowei Guo, Guangming Lu, Xing Sun.
paper code

Local to Global Learning: Gradually Adding Classes for Training Deep Neural Networks (CVPR 2019)
Hao Cheng, Dongze Lian, Bowen Deng, Shenghua Gao, Tao Tan, Yanlin Geng.
paper code

Evaluating Capability of Deep Neural Networks for Image Classification via Information Plane (ECCV 2018)
Hao Cheng, Dongze Lian, Shenghua Gao, Yanlin Geng.
paper code

Services

Reviewer of NeurlPS/ICLR/ICML/CVPR (2022-2026)

Student organizer of IJCAI 2022 workshop on 1st Learning and Mining with Noisy Labels Challenge

Speaker on IJCAI 2023 Tutorial: A Hands-on Tutorial for Learning with Noisy Labels