Zifeng Wang

I am a research scientist at Google. I received my PhD in Machine Learning at SPIRAL Group from Northeastern University, advised by Prof. Jennifer G. Dy. During my PhD, I also work closely with Prof. Stratis Ioannidis and Prof. Yanzhi Wang. I received my BS degree in Electronic Engineering from Tsinghua University. In my college years, I also worked with Prof. Jiwen Lu (Tsinghua), Prof. Jia Deng (Princeton) on computer vision and Prof. Yong Li (Tsinghua) on big data.

I am looking for self-motivated student researchers / research interns with interests and expertise in LLMs! Feel free contact me if you would like to work on cutting-edge challenges in machine learning and LLMs.

Email  /  Google Scholar  /  LinkedIn  /  Twitter  /  Resume  /  CV

• Large language model (LLM) Alignment
• Continual (Lifelong) learning
• Open set recognition / novel class discovery
• Adversarial robustness and model compression
• Deep learning applications in computer vision, natural language understanding, document understanding, healthcare, wireless communications, etc.

• 07/2023: I will be giving a talk on effective and efficient continual learning at A*STAR , thanks for hosting!
• 05/2023: Our paper on document entity extraction is accepted at ACL 2023 (Findings)
• 04/2022: I received the Outstanding Student Award in Research from COE, Northeastern University
• 04/2023: Our paper on continual learning is accepted at ICML 2023 , code will be released soon!
• 02/2023: I am glad to give talks about efficient and sparse continual learning at ContinualAI and AI Time
• 10/2022: I received the Scholar Award for NeurIPS 2022
• 09/2022: Our paper on efficient continual learning is accepted at NeurIPS 2022
• 09/2022: I am glad to give a talk about prompting-based continual learning at ContinualAI
• 09/2022: Our paper on adversarially robust pruning is accepted at ICDM 2022
• 07/2022: Our paper on prompting-based continual learning is accepted at ECCV 2022
• 03/2022: Our paper on prompting-based continual learning is accepted at CVPR 2022
• Earlier news
02/2022: Our paper on Bayesian continual learning is accepted by Neural Networks
09/2021: Our paper on HSIC for adversarial robustness is accepted at NeurIPS 2021
09/2021: Our paper on deep learning for smoking prediction is accepted by PLOS Computational Biology
05/2021: I started my research internship at Google Cloud AI Research
09/2020: Our paper on feature grouping is accepted at NeurIPS 2021
08/2020: Two papers on continual learning and open-world class discovery are accepted at ICDM 2020 , with one selected as the best paper candidate
08/2019: One paper on unseen class detection is accepted at DySPAN 2019 as the best paper

• 
  Aug 2023 - Present, Google,
  Research Scientist at Cloud AI Research
• 
  Sep 2018 - July 2023, Northeastern University,
  Research assistant at SPIRAL Group
• 
  June 2021 - Jan 2023, Google,
  Student researcher / Research Intern at Cloud AI Research
• 
  Feb 2017 - July 2018, Tsinghua University,
  Research assistant at i-Vision Group
• 
  July 2017 - Sep 2017, University of Michigan,
  Visiting researcher at Vision & Learning Lab

CHAIN-OF-TABLE enhances the reasoning capability of LLMs by leveraging tabular structures to express intermediate thoughts for table-based reasoning. It instructs LLMs to dynamically plan an operation chain according to the input table and its associated question.

QueryForm consists of a novel prompting-based framework for zero-shot document entity recognition with large language models (LLMs), and a large-scale weakly-supervised pre-training method on publicly available webpages.

DualHSIC consists of two complementary components that stem from the Hilbert Schmidt independence criterion (HSIC): HSIC-Bottleneck for Rehearsal (HBR) lessens the inter-task interference and HSIC Alignment (HA) promotes task-invariant knowledge sharing.

SparCL explores sparsity for efficient continual learning and achieves both training acceleration and accuracy preservation through the synergy of three aspects: weight sparsity, data efficiency, and gradient sparsity.

DualPrompt presents a novel approach to attach complementary prompts to the pre-trained backbone, and then formulates the continual learning objective as learning task-invariant and task-specific “instructions".

We propose a new learning paradigm for continual learning: our method learns to dynamically prompt (L2P) a pre-trained model to learn tasks sequentially under different task transitions.

We develop a fully Bayesian inference framework for ULL with a novel end-to-end Deep Bayesian Unsupervised Lifelong Learning (DBULL) algorithm.

We investigate the HSIC (Hilbert-Schmidt independence criterion) bottleneck as a regularizer for learning an adversarially robust deep neural network classifier, both theoretically and empirically.

We propose a novel deep learning model to incorporate prior knowledge regarding the relationship of exons to transcript isoforms. We hypothesized that since smoking alters patterns of exon and isoform usage, greater predictive accuracy could be obtained by using exon and isoform-level quantifications to predict smoking status.

We propose a learn-prune-share (LPS) algorithm which addresses the challenges of catastrophic forgetting, parsimony, and knowledge reuse simultaneously.

We propose Class Discovery Kernel Network with Expansion (CD-KNet-Exp), a deep learning framework for open-world class discovery problem.

We formally define two types of redundancies using information theory: Representation and Relevant redundancies. We leverage these redundancies to design a formulation for instance-wise feature group discovery and reveal a theoretical guideline to help discover the appropriate number of groups.

We propose a novel approach that facilitates new class detection without retraining a neural network, and perform extensive analysis of the proposed model both in terms of model parameters and real-world datasets.

We propose a collaborative deep reinforcement learning (C-DRL) method for multi-object tracking. Specifically, we consider each object as an agent and track it via the prediction network, and seek the optimal tracked results by exploiting the collaborative interactions of different agents and environments via the decision network

• Sparse Continual Learning on the Edge
  - ContinualAI, 2023, Remote
  - AI Time, 2023, Remote (Chinese version)
• Prompting-based Continual Learning
  - AI Talks SG, 2022, Remote
  - ContinualAI, 2022, Remote [recording]
• Revisiting Hilbert-Schmidt Information Bottleneck for Adversarial Robustness
  - INFORMS Annual Meeting, 2022, Indianapolis, Indiana [abstract]
  - AI Time, 2022, Remote (Chinese version) [recording][blog]

• Outstanding Student Award in Research, Northeastern University, 2023
  
• Scholar Award, NeurIPS 2022
  
• Best Paper Candidate, ICDM 2020
  
• Best Paper Award, DySPAN 2019
  
• Travel Award, DySPAN 2019
  
• Travel Award, NeurIPS 2019
  
• Dean's Fellowship, Northeastern University, 2018
     Highest honor awarded to new PhD students for out standing academic background.
• Evergrande Scholarship, Tsinghua University, 2016
     Awarded to students with excellent academic performance, scientific potential and overall development.

Conference Reviewer: NeurIPS 21-23, ICML 21-23, ICLR 22-23, CVPR 22-23, ICCV 23, ACL ARR
PC Member: SDM 23
Journal Reviewer: TPAMI, TMLR, Neural Networks