Home

The LTI Student Research Symposium (SRS) is a one-day series of talks and poster presentations designed both to increase awareness of the diverse aspects of language technologies research conducted by students within the LTI, as well as to introduce incoming LTI students to the work of current students.
One of the talks will be selected for a "Best Presentation Award", which will be announced at the end of the Symposium. The winner of the "Best Presentation Award" will receive a cash prize of $500. Additionally, two Honorable Mentions will be selected, each receiving a cash prize of $100, and two $100 for "Best Poster Award" will also be presented.
An oral presentation in the Symposium fulfills the yearly presentation requirement and the speaking requirement for LTI PhD and MS students.

 2013 Program Schedule
Wednesday, August 21, 2013
10:15am - 6:00pm
GHC 6115 

 
 TIME SPEAKER  TITLE 
  10:15AM 

Jaime Carbonell Opening Remarks  
 10:25AM William Wang  SRS 2013 Program Overview
 10:35AM
Session chair: Prasanna Kumar
Manaal Faruqui Multilinguality to the Rescue
11:00AM Siddharth Gopal Distributed Training of large-scale Logistic Models
 11:25PM Shashank Srivastava A Walk-based Semantically Enriched Tree Kernel Over Distributed Word Representations
 11:50PM Wenyi Wang  Usage Analysis of a Voice-based Social Media Service in India
 12:15PM Meghana Kshirsagar
Aaron Li
Ting-Hao Huang
Rohan Ramanath
Haohan Wang
Poster Madness  (A)
 12:30PM Lunch Break (food provided)
Poster Session (A)
GHC 6121
1:30PM
Session chair: Elijah Mayfield
Jun Araki Detecting Subevent Structure for Event Coreference Resolution
 1:55PM

 Zhou Yu  Automatic Prediction of Friendship via Multi-model Dyadic Features
 2:20PM  Bhavana Dalvi Exploratory Learning: Semi-supervised Learning in the Presence of Unanticipated Classes
 2:45PM Nathan Schneider Discriminative Lexical Semantic Segmentation with Gaps: Running the MWE Gamut
 3:10PM  Alok Kothari
Mrinmaya Sachan
Lara Martin
Adams Wei Yu
Matthew Marge
 Poster Madness  (B)
 3:25PM  Coffee Break / Poster Session (B) GHC 6121
 4:25PM
Session chair: William Wang
Keynote Speaker: John Kominek (Voci)  How LTI positively changed my life
 5:30PM
Session chair: Florian Metze
Awards / Closing Ceremony   

Poster Session (A)
  1. Meghana Kshirsagar. Multi-task learning for protein interaction prediction
  2. Aaron Li. Multi-GPU Distributed Parallel Bayesian Differential Topic Modelling
  3. Ting-Hao Huang. ACBiMA: Advanced Chinese Bi-Character Word Morphological Analyzer
  4. Rohan Ramanath. Crowd Prefers the Middle Path: A New IAA Metric for Crowdsourcing Reveals Turker Biases in Query Segmentation
  5. Haohan Wang. Behavior Analysis of Low-literate Users of Polly
Poster Session (B)
  1. Alok Kothari. Unsupervised Word Sense Disambiguation of Queries using Corewords
  2. Mrinmaya Sachan. A novel Paradigm to model Interactions and Content for detecting Compact Communities
  3. Lara Martin. Using Prosodic Cues from Uncertainty in Natural Speech to Help Computational Search
  4. Adams Wei Yu. Reverse Top-k Search using Random Walk with Restart
  5. Matthew Marge. Towards Evaluating Recovery Strategies for Situated Grounding Problems in Human-Robot Dialogue
Keynote Speaker: 
John
John Kominek (CTO, Vocitec Inc.)
John is a Voci co-founder, and brings the company successful start-up experience along with extensive expertise in speech technologies. In his first start-up, John was co-founder of the highly successful Quack.com, the first company to enter the voice portal business (and later acquired by AOL). At Quack, he led the engineering teams responsible for the fusion and structuring of voice-searchable data, and for the development of the text-to-speech architecture. Prior to Voci, John worked for Automated Speech Technologies, where he developed extremely low bitrate speech codecs based on vocabulary independent speech recognition. He has over 40 research publications and is the holder of five US patents. John earned his BS in Honours Physics and his MS in Mathematics and Computer Science at the University of Waterloo, and his PhD from the Language Technologies Institute at Carnegie Mellon.
Subpages (2): Contacts Goals