Zhe Liu | Research

DOC-RAG: ASR language model personalization with domain-distributed co-occurrence retrieval augmentation.
Puneet Mathur, Zhe Liu, Ke Li, Yingyi Ma, Gil Keren, Zeeshan Ahmed, Dinesh Manocha, and Xuedong Zhang. Joint International Conference on Computational Linguistics, Language Resources and Evaluation (LREC-COLING), 2024.

Correction focused language model training for speech recognition.
Yingyi Ma, Zhe Liu, and Ozlem Kalinli. IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2024.

Forgetting private textual sequences in language models via leave-one-out ensemble.
Zhe Liu and Ozlem Kalinli. IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2024.

Recovering from privacy-preserving masking with large language models.
Arpita Vats, Zhe Liu, Peng Su, Debjyoti Paul, Yingyi Ma, Yutong Pang, Zeeshan Ahmed, and Ozlem Kalinli. IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2024.

Contextual biasing of named-entities with large language models.
Chuanneng Sun, Zeeshan Ahmed, Yingyi Ma, Zhe Liu, Lucas Kabela, Yutong Pang, and Ozlem Kalinli. IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2024.

PersonaLM: language model personalization via domain-distributed span aggregated k-nearest n-gram retrieval augmentation.
Puneet Mathur, Zhe Liu, Ke Li, Yingyi Ma, Gil Keren, Zeeshan Ahmed, Dinesh Manocha, and Xuedong Zhang. Findings of the Association for Computational Linguistics: Empirical Methods in Natural Language Processing (EMNLP), 2023.

Modeling dependent structure for utterances in ASR evaluation.
Zhe Liu and Fuchun Peng. Annual Conference of the International Speech Communication Association (INTERSPEECH), 2023.

Adaptive multi-corpora language model training for speech recognition.
Yingyi Ma, Zhe Liu, and Xuedong Zhang. IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2023.

Mitigating unintended memorization in language models via alternating teaching.
Zhe Liu, Xuedong Zhang, and Fuchun Peng. IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2023.

Group personalized federated learning.
Zhe Liu, Yue Hui, and Fuchun Peng. IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2023.

FedSynth: gradient compression via synthetic data in federated learning.
Shengyuan Hu, Jack Goetz, Kshitiz Malik, Hongyuan Zhan, Zhe Liu, and Yue Liu. Neural Information Processing Systems (NeurIPS) Workshop on Federated Learning, 2022.

Model-based approach for measuring the fairness in ASR.
Zhe Liu, Irina-Elena Veliche, and Fuchun Peng. IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2022.

Neural-FST class language model for end-to-end speech recognition.
Antoine Bruguier, Duc Le, Rohit Prabhavalkar, Dangna Li, Zhe Liu, Bo Wang, Eun Chang, Fuchun Peng, Ozlem Kalinli, and Michael L. Seltzer. IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2022.

Private language model adaptation for speech recognition.
Zhe Liu, Ke Li, Shreyan Bakshi, and Fuchun Peng. arXiv preprint arXiv, 2110.10026, 2021.

Federated marginal personalization for ASR rescoring.
Zhe Liu and Fuchun Peng. IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2021.

Statistical testing on ASR performance via blockwise bootstrap.
Zhe Liu and Fuchun Peng. Annual Conference of the International Speech Communication Association (INTERSPEECH), 2020.

An empirical study of transformer-based neural language model adaptation.
Ke Li, Zhe Liu, Tianxing He, Hongzhao Huang, Fuchun Peng, Daniel Povey, and Sanjeev Khudanpur. IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2020.

Improving n-gram language models with pre-trained deep transformer.
Yiren Wang, Hongzhao Huang, Zhe Liu, Yutong Pang, Yongqiang Wang, ChengXiang Zhai, and Fuchun Peng. arXiv preprint, arXiv:1911.10235, 2019.

Learning nonparametric forest graphical models with prior information.
Yuancheng Zhu, Zhe Liu, and Siqi Sun. International Conference on Artificial Intelligence and Statistics (AISTATS), 2017.

Bayesian model-averaged regularization for Gaussian graphical models.
Zhe Liu. Communications in Statistics - Simulation and Computation, 46(4):3213-3223, 2017.

Graphical exponential screening.
Zhe Liu. arXiv preprint, arXiv:1503.02698, 2015.

Blossom tree graphical models.
Zhe Liu and John Lafferty. Advances in Neural Information Processing Systems (NeurIPS), 2014.

Support vector machines with L1 penalty for detecting gene-gene interactions.
Yuanyuan Shen, Zhe Liu, and Jurg Ott. International Journal of Data Mining and Bioinformatics, 6(5):463-470, 2012.

Challenging false discovery rate: a partition test based on p-values in case-control association studies.
Jurg Ott, Zhe Liu, and Yuanyuan Shen. Human Heredity, 74(1):45-50, 2012.

Multilocus association mapping using generalized ridge logistic regression.
Zhe Liu, Yuanyuan Shen, and Jurg Ott. BMC Bioinformatics, 12:384, 2011.

Genome-wide association study identifies three new susceptibility loci for esophageal squamous-cell carcinoma in Chinese populations.
Chen Wu, Zhibin Hu, Zhonghu He, Weihua Jia, Feng Wang, Yifeng Zhou, Zhihua Liu, Qimin Zhan, Yu Liu, Dianke Yu, Kan Zhai, Jiang Chang, Yan Qiao, Guangfu Jin, Zhe Liu, Yuanyuan Shen, Chuanhai Guo, Jianhua Fu, Xiaoping Miao, Wen Tan, Hongbing Shen, Yang Ke, Yixin Zeng, Tangchun Wu, and Dongxin Lin. Nature Genetics, 43(7):679-684, 2011.

Genome-wide examination of genetic variants associated with response to platinum-based chemotherapy in patients with small-cell lung cancer.
Chen Wu, Binghe Xu, Peng Yuan, Jurg Ott, Yin Guan, Yu Liu, Zhe Liu, Yuanyuan Shen, Dianke Yu, and Dongxin Lin. Pharmacogenet Genomics, 20(6):389-395, 2010.

Systematic removal of outliers to reduce heterogeneity in case-control association studies.
Yuanyuan Shen, Zhe Liu, and Jurg Ott. Human Heredity, 70(4):227-231, 2010.

Publications and Preprints

Thesis