The goal of Southern California Natural Language Processing Symposium is to gather researchers from the southern California region with broad expertise in natural language processing. The symposium will provide a participatory environment where attendees from a variety of fields working on natural language processing can share and discuss their latest findings.

This year, the SoCal NLP Symposium will be hosted at the University of Southern California, and it will include invited talks from academia and industry, contributed work, poster presentations, and open discussion. We welcome all students, postdocs, and faculty members from universities in the region, including University of Southern California (USC), UC Los Angeles, UC Irvine, UC Santa Barbara, UC Riverside, Caltech, UC San Diego, and other schools to join us.

Important Dates

Submission deadline: July 8, 2019 (Mon @ 11:59 PM PDT)
Notification of acceptance: Aug 14, 2019 (Wed)
SoCal NLP Symposium: Sep. 13th, 2019 (Fri)

Symposium Details

Browse the updated schedule here and register for the symposium here.

Thanks to our sponsors, registration and food are free for all attendees. Vegetarian options will be provided.

For those presenting a poster at the symposium, we will be providing poster boards and clips to hang your poster. The poster board has three panels and is 48 wide x 36 tall, so it would be best if your poster were at most 48" wide x 36" tall (horizontally) or 24" wide x 36" tall (vertically).

For those presenting a contributed talk at the symposium, you will have 15 minutes for the talk + questions.


Date: Friday, September 13th, 2019

Location: USC Caruso Catholic Center (Newman Hall). Please enter from the University Avenue entrance.

Parking information: You can park at the USC Village Parking Garage at 3131 S. Hoover Street, access to the garage is on the west side of the intersection of W. 32nd Street and S. Hoover Street. You can also park at the parking garage on the USC campus at S. Figueroa Street and USC McCarthy Way. Parking for the whole day is typically $10-$15, and unfortunately will not be reimbursed.


8:30am - 9:00am  Breakfast & registration
9:00am - 9:15am  Opening Remarks
9:15am - 9:50am  Invited Talk by Kevin Knight (Didi Chuxing) [Details]
9:50am - 10:25am  Invited Talk by Sameer Singh (UCI) [Details]
10:25am - 11:00am  Invited Talk by Liang Huang (Baidu / Oregon State University) [Details]
11:00am - 11:15am  Contributed Talk: Examining Gender Bias in Languages with Grammatical Genders
11:15am - 11:30am  Contributed Talk: Implicit Discourse Relation Identification for Open-domain Dialogue
11:30am - 1:00pm  Lunch & Poster Session #1
1:00pm - 1:35pm  Invited Talk by Marjan Ghazvininejad (Facebook) [Details]
1:35pm - 2:10pm  Invited Talk by Ndapa Nakashole (UCSD) [Details]
2:10pm - 2:25pm  Contributed Talk: Investigating Robustness and Interpretability of Link Prediction via Adversarial Modifications
2:25pm - 2:40pm  Contributed Talk: Dynamic Knowledge-Grounded Dialogue Generation through Walking on the Graph
2:40pm - 4:20pm  Coffee Break & Poster Session #2
4:20pm - 4:55pm  Invited Talk by Anima Anandkumar (Caltech) [Details]
4:55pm - 5:30pm  Invited Talk by Luke Zettlemoyer (University of Washington) [Details]
5:30pm - 5:45pm  Closing Remarks & Awards

Invited Speakers

Animashree Anandkumar

Bren Professor

California Institute of Technology

Marjan Ghazvininejad

Research Scientist

Facebook AI Research

Liang Huang

Distinguished Scientist / Assistant Professor

Baidu Research USA / Oregon State University

Kevin Knight

Chief Scientist for Natural Language Processing

Didi Chuxing

Ndapa Nakashole

Assistant Professor

Department of Computer Science & Engineering

University of California, San Diego

Sameer Singh

Assistant Professor

Department of Computer Science

University of California, Irvine

Luke Zettlemoyer

Associate Professor

Allen School of Computer Science & Engineering

University of Washington

Accepted Work

Contributed Talks

  • Examining Gender Bias in Languages with Grammatical Genders.
    Pei Zhou, Weijia Shi, Jieyu Zhao, Kuan-Hao Huang, Muhao Chen and Kai-Wei Chang.
  • Implicit Discourse Relation Identification for Open-domain Dialogue.
    Mingyu Derek Ma, Kevin K. Bowden, Jiaqi Wu, Wen Cui and Marilyn Walker.
  • Investigating Robustness and Interpretability of Link Prediction via Adversarial Modifications.
    Pouya Pezeshkpour, Yifan Tian and Sameer Singh.
  • Dynamic Knowledge-Grounded Dialogue Generation through Walking on the Graph.
    Yi-Lin Tuan, Yun-Nung Chen and Hung-Yi Lee.
Poster Session #1

  1. Logic Attention Based Neighborhood Aggregation for Inductive Knowledge Graph Embedding.
    Peifeng Wang.
  2. How Much Tech in FinTech? Evidence from Initial Coin Offerings.
    Yukun Liu, Jinfei Sheng and Wanyi Wang.
  3. Anonymizing SMS Data for Analysis of Interpersonal Relationships through NLP.
    Adriana Knight, Daniel Yang and Julie Medero.
  4. Highlights of Attention Mechanisms for Model Interpretability.
    Khalil Mrini, Franck Dernoncourt, Trung Bui, Walter Chang and Ndapa Nakashole.
  5. Dialog Data Collection for an Interactive Plotting Agent.
    Yutong Shao and Ndapa Nakashole.
  6. Discourse Tagging for Scientific Evidence Extraction.
    Xiangci Li, Gully Burns and Nanyun Peng.
  7. Efficient Contextual Representation Learning With Continuous Output Embedding.
    Liunian Harold Li, Patrick H. Chen, Cho-Jui Hsieh and Kai-Wei Chang.
  8. Reprogramming the Inner Critic: An Artistic Research Approach to NLP (and Human) Error.
    Sarah Ciston.
  9. Automatic Discovery of Language Dialects via Explainable Machine Learning.
    Amir Feghahati and Mike Izbicki.
  10. Reading KITTY: Pitch Range as an Indicator of Reading Skill.
    Alredo Gomez, Alessandra Otondo, Alicia Ngo and Julie Medero .
  11. Robust Abstractive Summarization via Deep Reinforcement Learning with Distributional Semantic Rewards.
    Siyao Li, Deren Lei, Pengda Qin and William Wang.
  12. Contextual Understanding of Homicide Reports in Los Angeles County.
    Sungyong Seo, Umang Gupta, Jiageng Zhu, P. Jeffrey Brantingham and Yan Liu.
  13. Target Language-Aware Constrained Inference for Cross-lingual Dependency Parsing.
    Tao Meng, Nanyun Peng and Kai-Wei Chang.
  14. Zero Shot Sentiment Analysis on Tweets in Any Language.
    Oscar Hernandez and Mike Izbicki.
  15. Towards Healthy Social Media Conversation in Low-Resource Languages.
    Samar Haider, Luca Luceri, Ashok Deb, Adam Badawy, Nanyun Peng and Emilio Ferrara.
Poster Session #2

  1. Building a Better Lie Detector with BERT: A First Step to Finding the Rules of Deception.
    Dan Barsever, Emre Neftci and Sameer Singh.
  2. A Hierarchical Training Method for User Geolocation.
    Sharon Levy, Elizabeth Belding and William Yang Wang.
  3. Cooking Common Sense: Personalized Recipe `Tweak' Inference via Common Sense Reasoning.
    Shuyang Li, Bodhisattwa Majumder and Julian McAuley.
  4. Challenges of Applying Sentiment Analysis on Health-related Social Media Data.
    Tingjue Yin and Lu He.
  5. Word Vectors for 244 Countries from Tweets for 300 Spanish Dialects Using Factored Multiskipgram Model.
    Rany Tith and Mike Izbicki .
  6. Distilling Pre-Trained Language Models into Compact Graph-Augmented Networks.
    Bodhisattwa Prasad Majumder, Henry Mao and Julian McAuley.
  7. Developing a Corpus of Descriptions of Personal Struggles.
    Sheridan Stewart and Jeremy Freese.
  8. Integrating Local Structure into Knowledge Graph Embeddings.
    Pouya Pezeshkpour, Yifan Tian and Sameer Singh.
  9. Improving Generalization in Coreference Resolution via Adversarial Training.
    Sanjay Subramanian and Dan Roth.
  10. Using Knowledge Graphs for Fact-Aware Language Modeling.
    Robert Logan, Sameer Singh, Matt Gardner, Matthew Peters and Nelson Liu.
  11. Effective Forum Curation via Multi-task Learning.
    Faeze Brahman, Nikhil Varghese, Snigdha Chaturvedi and Suma Bhat.
  12. Automatic Evaluation of Open-domain Dialogue Systems on Relevancy and Engagement.
    Sarik Ghazarian, Aram Galstyan and Nanyun Peng.
  13. Reestablishing Trust When Communications Between Human Users and Virtual Agents Go Awry.
    Patricia Chaffey.
  14. Modeling "Newsworthiness" for Lead-Generation.
    Alexander Spangher.
  15. The First Casualty of War is Truth: A Dataset for Multilingual Fake News Detection.
    Samar Haider, Ashok Deb, Nanyun Peng and Emilio Ferrara.



Nanyun (Violet) Peng

Research Assistant Professor

Department of Computer Science


Jonathan May

Research Assistant Professor

Department of Computer Science


Student Co-Chairs

Rujun (RJ) Han

Ph.D. Student

Department of Computer Science


Meryem M'hamdi

Ph.D. Student

Department of Computer Science


Emily Sheng

Ph.D. Student

Department of Computer Science


Past Symposiums


If you need accommodations or have questions or comments, please email us at socalnlp@googlegroups.com.