Resume

Academic & Professional Background

Download PDF

Highlights

Professional Experience

Research Scientist Sep 2016 - Oct 2018
R&D Manager Oct 2018 - Oct 2022
Senior ML Research Scientist Oct 2022 - present
Apple, Cupertino, CA
  • Lead the prototyping and development of applications natural language interactions, code generation and UI navigation, leveraging LLMs and modern ML.
  • Investigated Web-based knowledge graph expansion of Siri's internal facts for QA over knowledge domains.
  • Lead a team of applied researchers to build voice-driven query understanding models for on-device deployment (model compression, model development).
  • Prototyped methods for scaling natural language understanding in Siri to over 20 worldwide languages.
  • Developed scalable models for knowledge-graph question answering in Siri, with international scale-up across supported locales.
Post-Doctoral Researcher Oct 2009 - Feb 2012
Senior Software Engineer Feb 2012 - Nov 2014
Research Staff Member Nov 2014 - Aug 2016
IBM T. J. Watson Research Center, Yorktown Heights, NY
  • Developed answer scoring and question analysis components for IBM Watson -- a ground-breaking question answering system that defeated grand champions Ken Jennings and Brad Rutter in a nationally televised challenge on the TV game show Jeopardy! [in JNLE, IBM Journal of R&D, CIKM '12, EMNLP '11, CIKM '11]
  • Researched "scenario-based" question answering in the medical domain with IBM Watson, through inference "chaining" over unstructured data. [in AI Magazine '17, IBM Research Report]
  • Conducted research in question answering on electronic health records. [in ACL '16, and podium abstract in AMIA-CRI '16]
  • Built event-focused information extraction for automatic "machine reading" and reasoning for complex Q&A over text documents (DARPA funded). [in EACL '12 and COLING '10]
Summer Research Intern May 2007 - Aug 2007
IBM T. J. Watson Research Center, Yorktown Heights, NY
  • Conducted research on automatic speech recognition models, through an analysis of errors in recognizing Keywords and reducing Word Error Rate. [in Interspeech '08]
Research Assistant Aug 2003 - Oct 2009
School of Computing, University of Utah, Salt Lake City, UT
  • Conducted research on probabilistic event-focused IE, using SVM and CRFs with textual features in event descriptions. [in EMNLP '09]
  • Designed and implemented two-phase SVM-based model for event information extraction from text through text-based features. [in EMNLP-CoNLL '07]
  • Developed an approach for greater coverage of text-based IE with new lexico-syntactic pattern-based approaches. [in ACL '06 workshop]
  • Implemented a hierarchical structure (feature subsumption) for selecting the best ML features in opinion detection in text. [in EMNLP '06]
  • Designed and implemented algorithms (WordNet::Similarity) for automatic measurement of the semantic similarity and relatedness of English words and concepts. [system demo at NAACL '04 and AAAI '04]
  • Created and open-sourced system (WordNet::SenseRelate) for disambiguating the intended meaning of any given word from its context, using the notion of semantic relatedness of concepts. [system demo at ACL '05 and AAAI '05]
Summer Research Intern May 2003 - Aug 2003
Division of Biomedical Informatics, Mayo Clinic, Rochester, MN
  • Adapted WordNet-based approaches for measuring semantic similarity to applications in the biomedical domain; built open source modules Snomed::Interface, Semantic::Similarity and UMLS::Similarity for the biomed domain. [in JBI]

Education

PhD, Computer Science May 2010
University of Utah
MS, Computer Science Aug 2003
University of Minnesota
BE, Computer Engineering May 2001
University of Pune

Professional Service & Awards

Awards
  • Outstanding Technical Contribution Award, IBM Watson Group (2014)
  • AAAI Feigenbaum Prize (to the IBM Watson team), the Association for Advancement of Artificial Intelligence (2013)
Area Chair
  • EMNLP '25, ACL '25, NAACL '25, AAAI '25, EMNLP '24, ACL '24, NAACL '24, EACL '24, AAAI '24, AACL '22, NAACL Industry Track '22, ACL '21, ACL '19, COLING '14
Senior Program Committee
  • AAAI '26, AAAI '23, AAAI '22, AAAI '21, IJCAI '16
Publications Chair
  • EMNLP '17, EMNLP '16
Advisory Boards
  • DropThought Inc. (2011 - 2016)
  • Scientific Advisory Board, NIST TAC-KBP 2020
Program Committee
  • EMNLP Industry Track '25, COLM '24, EMNLP '23, BlackboxNLP Workshop '23, ACL '23, EMNLP '22, COLING '22, NAACL '22, ACL'22, BlackboxNLP Workshop '21, EMNLP '21, NAACL '21, EMNLP '20, COLING '20, IJCAI '20, ACL '20, WeCNLP '19, EMNLP '19, NAACL '19, IJCAI '19, AAAI '19, EMNLP '18, ACL '18, NAACL '18, COLING '18, IJCAI-ECAI '18, EMNLP '17, ACL '17, EACL '17, AAAI '17, COLING '16, IJCAI '16, NAACL '16, ACL '16, AMIA CRI '16, *SEM '16, EMNLP '15, ACL-IJCNLP '15, NAACL '15, *SEM '15, EMNLP '14, ACL '14, EACL '14, LREC '14, SIGIR '14 Workshop on Semantic Matching in Information Retrieval, IJCNLP '13, ACL '13, NAACL-HLT '13, COLING '12, EMNLP-CoNLL '12, ACL '12, NAACL-HLT '12, RANLP '11 Workshop on Information Extraction and Knowledge Acquisition, EMNLP '11, ACL-HLT '11, EMNLP '10, COLING '10, NAACL-HLT '10, ACL '08
Reviewer
  • Transactions on Interactive Intelligent Systems ('25), Computational Linguistics ('23, '22, '21, '20), Natural Language Engineering ('23, '18, '15, '13), Computer Speech & Language ('22), Transactions of the Association for Computational Linguistics ('21, '20, '19, '18, '17, '16, '15, '14, '13), IEEE Transactions on Audio, Speech and Language Processing ('17, '16), Artificial Intelligence ('15), Language Resources and Evaluation ('12), Data Mining and Knowledge Discovery ('11), Journal of Artificial Intelligence Research ('09), NESCAI ('08, '07)
Auxiliary Reviewer
  • AAAI '16, CIKM '14, AAAI '06
Session Chair
  • EMNLP '24, CL '21, ACL '19, CIKM '12, ACL '06 Workshop on IE Beyond the Document
Other
  • Mentor, AI Generator '21, AI@MIT
  • Graduate Student Advisory Committee, School of Computing, University of Utah (2004)
  • Most Outstanding Teaching Assistant, University of Minnesota Duluth, MN (2003)
  • National Talent Search Scholarship, National Council for Educational Research & Training, India (1995 - 2001)

Patents

Entailment Knowledge Base in Natural Language Processing Systems
B. Boguraev, J. Chu-Carroll, A. Kalyanpur, D. McClosky, J. W. Murdock and S. Patwardhan
US Patent 11,520,813 B2 and US Patent 11,720,611 B2, Aug 2023
Adaptable Processing Components
Y. Park and S. Patwardhan
US Patent 11,144,718 B2, October 2021
Tone Analysis of Legal Documents
H. Badenes, R. Mannan and S. Patwardhan
US Patent 10,452,780, and US Patent 10,929,615, February 2021
Processing Context-based Inquiries for Knowledge Retrieval
S. Bagchi, K. Barker, B. Boguraev, M. Bornea, A. Faulkner, Y. Li, S. Patwardhan, and S. Rosenthal
US Patent 10,769,138, September 2020
Self-Training of Question Answering System using Question Profiles
M. Devarakonda, S. Patwardhan and P. Raghavan
US Patent 10,699,215, June 2020
Generating Secondary Questions in an Introspective Question Answering System
B. Boguraev, D. Buchanan, J. Chu-Carroll, D. Ferrucci, A. Kalyanpur, J. W. Murdock, and S. Patwardhan
US Patent 10,621,880, US Patent 10,614,725, April 2020
Annotation of Legal Documents with Case Citations
H. Badenes, R. Mannan and S. Patwardhan
US Patent 10,373,278, August 2019
Automated Timeline Completion Using Event Progression Knowledge Base
M. Devarakonda, S. Patwardhan and P. Raghavan
US Patent 10,372,822, August 2019
Robust and Readily Domain-Adaptable Natural Language Interface to Databases
B. Boguraev, E. Khorasani, V. Sheinin, S. Patwardhan and P. Zerfos
US Patent 10,042,921 B2, August 2018
Natural Language Interface to Databases
B. Boguraev, E. Khorasani, V. Sheinin, S. Patwardhan and P. Zerfos
US Patent 9,959,311 B2, May 2018
Natural Language Interface to Databases
B. Boguraev, E. Khorasani, V. Sheinin, S. Patwardhan and P. Zerfos
WO Patent WO/2017/046729, March 2017
Collaborative Creation of Annotated Training Data
S. Berajawala, A. Levas, S. Patwardhan, and D. Taieb
US Patent 9,860,308 B2, January 2018
Fact Checking Using and Aiding Probabilistic Question Answering
D. Ferrucci, D. Gondek, A. Kalyanpur, A. Lally, and S. Patwardhan
US Patent 8,972,321 B2 and US Patent 8,959,043 B2, February 2015
Adaptable Processing Components
Y. Park and S. Patwardhan
WO Patent Application WO/2018/158626, filed December 2017
Patent Pending
Identification of Related Electronic Medical Record Documents in a Question and Answer System
M. Devarakonda, J. Liang, S. Patwardhan and P. Raghavan
US Patent Application 15/336,261, filed October 2016
Patent Pending

Invited Presentations & Panels

Discussion Panel on IE/IR Feb 2021
Invited Panelist, TAC-KBP 2020 Workshop
Semantic Technologies for Automatic Question Answering Jul 2016
Visiting Professor, ECI 2016 Winter School, University of Buenos Aires, Argentina
Watson: From Jeopardy! to Medicine Jul 2016
Invited Talk, IBM, Argentina
Cognitive Computing: Question Answering Technologies Behind Watson Mar 2015
Guest Instructor, IBM Online Course, IBM Watson Academy, IBM, Armonk NY
Case Study on IBM Watson Nov 2015
Invited Guest Commentator, MBA Course: "Technology and Operations Management", Harvard Business School, Boston MA
UIMA in IBM Watson Oct 2015
Guest Lecturer, Computer Science Course: "Semantic Technologies in IBM Watson", Columbia University, New York NY
Building Applications with Watson Experience Manager Oct 2014
Guest Lecturer, Computer Science Course: "Natural Language Processing with Watson", Rensselaer Polytechnic Institute, Troy NY
Building IBM's Watson: Technical Insights into the Jeopardy-winning DeepQA Technology Sept 2014
Invited Speaker, First Workshop on Cognitive Systems and Watson, National University of Singapore, Singapore
Jeopardy! "Special" Questions: Exploring the Boundaries of Watson March 2014
Guest Lecturer, Computer Science Course: "Natural Language Processing with Watson", Rensselaer Polytechnic Institute, Troy NY
Watson Beyond Jeopardy: Adaptation to the Medical Domain March 2014
Invited Speaker, Computer Science Colloquium, Rensselaer Polytechnic Institute, Troy NY
UIMA in IBM Watson March 2014
Guest Lecturer, Computer Science Course: "Semantic Technologies in IBM Watson", Columbia University, New York NY
UIMA in IBM Watson March 2013
Guest Lecturer, Computer Science Course: "Semantic Technologies in IBM Watson", Columbia University, New York NY
Building Watson: An Overview of the DeepQA Project for the Jeopardy! Challenge Feb 2012
Invited Speaker, Computer Science Colloquium, University of Utah, Salt Lake City UT
Building Watson: An Overview of the DeepQA Project for the Jeopardy! Challenge Oct 2011
Invited Speaker, External Speaker Series, Knight Capital Group, Jersey City NJ

Teaching

Instructor, 'Teach the Teacher' Course Jul 2016
IBM, Argentina
Visiting Professor, Semantic Technologies for Automatic Question Answering Jul 2016
ECI 2016 Winter School, University of Buenos Aires
Guest Instructor, Cognitive Computing: Question Answering Technologies Behind Watson Mar 2015
IBM Online Course: http://ibm.com/watsonacademy
Teaching Assistant, Machine Learning Fall 2003
School of Computing, University of Utah
Teaching Assistant, Operating Systems Practicum Spring 2003
Department of Computer Science, University of Minnesota Duluth
Teaching Assistant, Introduction to Natural Language Processing Fall 2002
Department of Computer Science, University of Minnesota Duluth
Teaching Assistant, Computer Science II Spring 2002
Department of Computer Science, University of Minnesota Duluth
Teaching Assistant, Computer Science I Fall 2001
Department of Computer Science, University of Minnesota Duluth

Selected Publications

EELBERT: Tiny Models through Dynamic Embeddings
G. Cohn, R. Agarwal, D. Gupta, and S. Patwardhan
Proceedings of EMNLP 2023 Industry Track, Singapore, December 2023
Languages You Know Influence Those You Learn: Impact of Language Characteristics on Multi-Lingual Text-to-Text Transfer
B. Muller, D. Gupta, J. Fauconnier, S. Patwardhan, D. Vandyke, S. Agarwal
NeurIPS 2022 Workshop on Transfer Learning for Natural Language Processing, New Orleans, LA, December 2022
Model Stability with Continuous Data Updates
H. Liu, A. Polisetty, S. Patwardhan, P. Grasch, and S. Agarwal
Preprint arXiv: 2201.05692, January 2022
Can Open Domain Question Answering Systems Answer Visual Knowledge Questions
J. Zhang, A. Mishra, A. Polisetty, S. Patwardhan, and S. Agarwal
Preprint arXiv: 2202.04306, February 2022
WatsonPaths: Scenario-based Question Answering and Inference over Unstructured Information
A. Lally, S. Bagchi, M. Barborak, D. Buchanan, J. Chu-Carroll, D. Ferrucci, M. Glass, A. Kalyanpur, E. Mueller, J. W. Murdock, S. Patwardhan, and J. Prager
AI Magazine, Summer 2017
Addressing Limited Data for Textual Entailment across Domains
C. Shivade, P. Raghavan, and S. Patwardhan
Proceedings of the 54th Annual Meeting of the Association for Computational Linguistics, Berlin, Germany, August 2016
[acceptance rate 28%]
The Role of Context Types and Dimensionality in Learning Word Embeddings
O. Melamud, D. McClosky, S. Patwardhan, and M. Bansal
Proceedings of NAACL-HLT 2016, San Diego, CA, June 2016
[acceptance rate 25%]
Parallel and Nested Decomposition for Factoid Questions
B. Boguraev, S. Patwardhan, A. Kalyanpur, J. Chu-Carroll, and A. Lally
Natural Language Engineering, October 2014
Fact-based Question Decomposition in DeepQA
A. Kalyanpur, S. Patwardhan, B. Boguraev, A. Lally, and J. Chu-Carroll
IBM Journal of Research and Development, May/July 2012
Structured Data and Inference in DeepQA
A. Kalyanpur, B. Boguraev, S. Patwardhan, J. W. Murdock, A. Lally, C. Welty, J. Prager, B. Coppola, A. Fokoue-Nkoutche, L. Zhang, Y. Pan, and Z. Qiu
IBM Journal of Research and Development, May/July 2012
Question Analysis: How Watson Reads a Clue
A. Lally, J. Prager, M. McCord, B. Boguraev, S. Patwardhan, J. Fan, P. Fodor, and J. Chu-Carroll
IBM Journal of Research and Development, May/July 2012
Using Syntactic and Semantic Structural Kernels for Classifying Definition Questions in Jeopardy!
A. Moschitti, J. Chu-Carroll, S. Patwardhan, J. Fan, and G. Riccardi
Proceedings of the 2011 Conference on Empirical Methods in Natural Language Processing, Edinburgh, UK, July 2011
[acceptance rate 23%]
Feature Subsumption for Opinion Analysis
E. Riloff, S. Patwardhan, and J. Wiebe
Proceedings of the 2006 Conference on Empirical Methods in Natural Language Processing, Sydney, Australia, July 2006
[acceptance rate 31%]
Identifying Sources of Opinions with Conditional Random Fields and Extraction Patterns
Y. Choi, C. Cardie, E. Riloff, and S. Patwardhan
Proceedings of Human Language Technology Conference and Conference on Empirical Methods in Natural Language Processing, Vancouver, Canada, October 2005
[acceptance rate 32%]