Midwest Speech and Language Days &
Midwest Computational Linguistics Colloquium 2017
May 4-5, 2017
Toyota Technological Institute at Chicago
6045 S. Kenwood Ave, Chicago, IL

Midwest Speech and Language Days (MSLD) & the Midwest Computational Linguistics Colloquium (MCLC) is a 2-day meeting that continues and expands upon the tradition of Illinois Speech Day and the Midwest Computational Linguistics Colloquium. Presenters and attendees come from Midwest universities and research institutions. The goal is to increase awareness of speech and language research going on in the region and to foster collaboration among sites.

This year MSLD & MCLC was held May 4-5, 2017 at TTI-Chicago. Thanks to all who participated! The next MSLD will be in spring 2018.



Sanjeev Khudanpur, Johns Hopkins University - Innovations in Acoustic Modeling for Automatic Speech Recognition


Kevin Gimpel, TTIC
Karen Livescu, TTIC
Liang Lu, TTIC


Talks: Talk slots are 20 minutes and include 15 minutes for the talk and 5 for questions.
Posters: Posters can be at most 4 feet by 6 feet. Please see your poster number below (day 1 and day 2 are different poster sets) and put your poster up at the corresponding number tag on the 5th floor walls.
Parking: There is free parking in the commuter parking lot at 60th St and Stony Island Ave and free street parking on many streets near TTIC (just beware of "permit parking" and "street cleaning" signs!). Parking can be found on 61st Street (between Woodlawn Ave and Blackstone Ave) and on Dorchester Street (between 60th and 61st Streets).
Accommodations: For accommodations we recommend La Quinta (4900A S. Lake Shore Drive Chicago, IL 60615), which also has a shuttle to UChicago, or the Hyatt Place Chicago-South. There are also many reasonable options in downtown Chicago, such as Hotel Felix.


Day 1: Thursday, May 4

10:00-10:55    Breakfast (catered at TTIC), hang Day 1 posters

10:55-11:00    Opening remarks

11:00-12:00    Keynote: Sanjeev Khudanpur

12:00-2:00    Lunch (catered at TTIC) + poster session

2:00-3:20    Talk session
2:00-2:20     Jonathan Kummerfeld (UMichigan) - Modeling Discourse Structure with Conversation Graphs
2:20-2:40     Camille Goudeseune and Mark Hasegawa-Johnson (UIUC) - Topic labeling of speech in low-resource languages
2:40-3:00     John Wieting and Kevin Gimpel (TTIC) - Learning Paraphrastic Representations: New Models and Sentential Paraphrase Generation
3:00-3:20     Jacob B. Phillips and Alan C. L. Yu (UChicago) - A longitudinal examination of individual variation in sibilant coarticulation and change

3:20-4:30    Break + poster session

4:30-5:30    Talk session
4:30-4:50     Shubham Toshniwal, Hao Tang, Liang Lu, and Karen Livescu (TTIC) - Multitask Learning with Low-Level Auxiliary Tasks for Encoder-Decoder Based Speech Recognition
4:50-5:10     David L. King, Andrea Sims, and Micha Elsner (OSU) - Neural Sequence to Sequence Learning for Russian Morphology
5:10-5:30     Mary Pietrowicz, Mark Hasegawa-Johnson, and Karrie Karahalios (UIUC) - Exploring Expressive Dimensions of Speech in Semi-structured, Unscripted Oral History Interviews

6:00    Optional group dinner at nearby restaurant; depart TTIC around 6pm

Day 1 Posters:
1. Mahmoud Azab and Mingzhe Wang (UMichigan) - Structured Matching for Phrase Localization
2. Deblin Bagchi and Eric Fosler-Lussier (OSU) - Attention-based models for segmental speech recognition
3. Wenda Chen (UIUC) - Mismatched Crowdsourcing From Multiple Annotator Languages For Recognizing Zero-resourced Languages: A Nullspace Clustering Approach
4. Zewei Chu, Hai Wang, Kevin Gimpel, and David McAllester (TTIC) - Broad Context Language Modeling as Reading Comprehension
5. Dmitriy Dligach, Timothy Miller, Chen Lin, Steven Bethard, and Guergana Savova (Loyola) - Neural Models for Clinical Temporal Relation Extraction
6. Eric Fosler-Lussier, Florian Metze, Rebecca Bates, and Eric Riebling (OSU) - The Speech Recognition Virtual Kitchen
7. Yanwen Mao, Micha Elsner, and Marie-Catherine de Marneffe (OSU) - Computational Analysis of Indirect Requests in English Business Emails: A Cross-Cultural Comparison among English Learners from Hong Kong , Japan and Native English Speakers
8. Denis R. Newman-Griffis, Albert M. Lai, and Eric Fosler-Lussier (OSU) - A Weakly-Supervised Method for Jointly Embedding Concepts, Phrases, and Words
9. Shane Settle, Keith Levin, Herman Kamper, and Karen Livescu (TTIC) - Query-by-Example Search with Discriminative Neural Acoustic Word Embeddings
10. Bowen Shi and Karen Livescu (TTIC) - A neural attention model for American Sign Language fingerspelling recognition
11. Lifu Tu and Kevin Gimpel (TTIC) - Adversarial Energy Structure Prediction Network
12. Liming Wang (UIUC) - Speech recognition for under-resourced language using HMM-based transfer learning approach
13. Zhao Wang and Aron Culotta (IIT) - Towards Quantifying the Authenticity of Public Messaging in Social Media
14. Laura Wendlandt, Rada Mihalcea, Ryan L. Boyd, and James W. Pennebaker (UMichigan) - Picture This: Multimodal Prediction and Analysis of Latent User Dimensions
15. Yihan Zhou (UIUC) - A computational approach to code switching points in Mandarin-English code switching corpus

Day 2: Friday, May 5

9:00-10:30    Breakfast (catered at TTIC) + poster session

10:30-11:50    Talk session
10:30-10:50     Daniel Khashabi, Tushar Khot, Ashish Sabharwal, and Dan Roth (UIUC) - Question Answering via Integer Programming over Semi-Structured Knowledge
10:50-11:10     Stephanie Antetomaso, Kouki Miyazawa, Naomi Feldman, Micha Elsner, Kasia Hitczenko, and Reiko Mazuka (OSU) - Modeling Phonetic Category Learning from Natural Acoustic Data
11:10-11:30     Yiyi Wang, Yanan Liu, and Chilin Shih (UIUC) - Automated Essay Scoring Using Deep Neural Networks
11:30-11:50     Herman Kamper, Shane Settle, Greg Shakhnarovich, and Karen Livescu (TTIC) - Visually grounded learning of keyword prediction from untranscribed speech

11:50-1:30    Lunch (catered at TTIC) + poster session

1:30-2:30    Talk session
1:30-1:50     Andrea F. Daniele (TTIC) - Navigational Instruction Generation as Inverse Reinforcement Learning with Neural Machine Translation
1:50-2:10     Doug Downey (NU) - Scaling-up Hierarchically-structured Word Representations
2:10-2:30     Micha Elsner and Cory Shain (OSU) - Word Segmentation with Neural Net Working Memory

Day 2 Posters:
1. Sabita Acharya, Barbara Di Eugenio, Andrew D. Boyd, Richard Cameron, Karen Dunn Lopez, Pamela Martyn-Nemeth, Carolyn Dickens, and Amer Ardati (UIC) - Towards personalizing hospital-stay summaries for patients
2. William Bryce, Ryan Cotterell, and Lane Schwartz (UIUC) - Dual Decomposition for Statistical Machine Translation
3. Zheng Cai, Lifu Tu, and Kevin Gimpel (TTIC) - The Value of Curated Negative Examples: Strong Baselines for Story Cloze Tasks
4. Lane Schwartz and Emily Chen (UIUC) - Liinnaqumalghiit - A Web-based Learning Framework for St. Lawrence Island / Central Siberian Yupik
5. Zhongtian Dai and Zheng Cai (TTIC) - Glyph-based Visual Chinese Character Embedding
6. Aparna Garimella, Rada Mihalcea, and James W. Pennebaker (UMichigan) - Identifying Cross-Cultural Differences in Word Usage
7. Itika Gupta, Barbara Di Eugenio, Brian Ziebart, Bing Liu, Ben Gerber, and Lisa Sharp (UIC) - Dialog-Based Virtual Assistant Health Coach for Improving Physical Activity using SMART Goal Setting
8. Mark Hasegawa-Johnson (UIUC) - Clustering Grapheme-to-Phoneme Transducers Across 100 Languages
9. Abhinav Kumar (UIC) - Conversational Interface for Data Visualization
10. Marten van Schijndel (OSU) - Approximations of Predictive Entropy Correlate with Reading Times
11. Adam Stiff and Eric Fosler-Lussier (OSU) - Spiking Networks for Phone Classification
12. Qingming Tang, Weiran Wang, and Karen Livescu (TTIC) - Acoustic feature learning with deep variational canonical correlation analysis
13. Charlie Welch, Song Feng, and Rada Mihalcea (UMichigan) - World Knowledge for Semantic Parsing with Abstract Meaning Representation
14. Michael White, Laura Arbogast, Evan Jaffe, and Douglas Danforth (OSU) - Combining Pattern Matching and Logistic Regression for Question Interpretation in a Virtual Patient Dialogue System
15. Steven Wilson and Rada Mihalcea (UMichigan) - Measuring Semantic Relations between Human Activity Phrase Pairs
16. Xiao Zhang, Maria Leonor Pacheco, Chang Li, and Dan Goldwasser (Purdue) - DRaiL: Towards Declarative Deep Relational Learning

Past events:

MSLD & MCLC 2016 at Indiana University
MSLD 2015 at TTI-Chicago
MSLD 2014 at UIUC
MSLD 2013 at TTI-Chicago
Illinois Speech Day 2012 at TTI-Chicago
Illinois Speech Day 2011 at TTI-Chicago
Illinois Speech Day 2010 at TTI-Chicago
Illinois Speech Day 2009 at TTI-Chicago
MCLC 2009 at Indiana University
MCLC 2008 at Michigan State University
MCLC 2006 at UIUC
MCLC 2005 at The Ohio State University
MCLC 2004 at Indiana University