Li "Harry" Zhang  张力

About Me

I am a first-year PhD student focusing on Natural Language Processing, working with Prof. Chris Callison-Burch at the University of Pennsylvania. I graduated from the University of Michigan in 2018, previously mentored by Prof. Rada Mihalcea and Prof. Dragomir Radev.

CV   简历   Publications      

University of Pennsylvania

NLPNatural Language Processing, MLMachine Learning, AIArtificial Intelligence

Shenzhen, China



I am a drummer and guitarist, occasionally making covers. I used to be in a band, but now I mostly practice, jam or record for the fun of it.


I am a competitive pool player. In college I was in the university team and played in intercollegiate tournaments. Now I mostly just train and practice.


I took a leap of faith from:

Research Highlights

Semi-Supervised Crosslingual Event and Entity Extraction
Sept 2019 - Present
This work is a part of the IARPA BETTER project. We are given training data of event and Entity extraction in English, and attemt to transfer the model to Arabic.

Split and Rephrase: Evaluation Benchmarks and Metrics
Apr 2019 - Jun 2019
Split and Rephrase is a text simplification task to rewrite a complex sentence into several simpler ones. We show that the existing benchmark is too simplistic, developing a rule-based model using no training data which performs on par with the current state-of-the-art neural model. We then propose two new crowdsourced benchmarks with improved quality. We also provide a study on the flaws of BLEU score, and the cost-efficiency of using crowd workers to evaluate.

Sentence Embeddings, Transfer Learning and Semantic Similarity [2] [3]
Oct 2017 - Sept 2018
Recent advancement on neural sentence embeddings show highly competitive performance on semantic similarity tasks. However, the embeddings don't usually just work off-the-shelf, as we show that the transfer learning methodology is crucial to performance. We propose a fine-tuning approach and a multi-label approach which outperforms most alternative transfer learning approaches on semantic similarity tasks, achieving state-of-the-art performance on multiple datasets.

Work and Teaching Experience

Research Intern @ IBM ResearchIBM Research
Apr 2019 - Jun 2019

I did NLP research and software development on text simplification, the Split and Rephrase task. See more in Research Highlights.

Summer Analyst in Technology @ Goldman SachsGoldman Sachs
May 2017 - Aug 2017

As a full-stack developer, I enhanced the GS App Store, the firm’s internal application delivery and management software. The technology stack for GS App Store consists of AngularJS, C# Web APIs and Elasticsearch. My major goal was to improve user experience using data analytics and machine learning.

Teaching AssistantUM-Penn
Sep 2016 - Present

At University of Pennsylvania, I instructed CIS 530: Computational Linguistics (Winter 2020). At University of Michigan, I instructed EECS 595: Natural Language Processing (Fall 2018) and EECS 280: Programming and Introductory Data Structures (Winter,Fall 2016).


[3] Multi-Label Transfer Learning for Multi-Relational Semantic Similarity
Li Zhang, Steven R. Wilson and Rada Mihalcea

Paper BibTeX Slides   In *SEM 2019; presented at NAACL 2019.

[2] Direct Network Transfer: Transfer Learning of Sentence Embeddings for Semantic Similarity
Li Zhang, Steven R. Wilson and Rada Mihalcea

Paper BibTeX Poster   In arXiv pre-print; presented at IC2S2 2018.

[1] Improving Text-to-SQL Evaluation Methodology
Catherine Finegan-Dollak, Jonathan K. Kummerfeld, Li Zhang, Karthik Ramanathan Dhanalakshmi Ramanathan, Sesh Sadasivam, Rui Zhang and Dragomir Radev

Paper BibTeX Code Poster   In ACL 2018.


University of Pennsylvaniaupenn logo
Sep 2019 - Present
Philadephia, PA, USA
Ph.D. Computer Science

In progress.

University of Michiganumich logo
Sep 2015 - Dec 2018
Ann Arbor, MI, USA
B.S.E. Computer Science

GPA: 3.82/4.00 summa cum laude

Shenzhen Middle SchoolSMS logo
Sep 2012 - Jun 2015
Shenzhen, China
High School Diploma

GPA: 4.23/4.30