Mohit Bansal

Ph.D. Candidate Computer Science Division University of California, Berkeley mbansal -atsign- cs -dot- berkeley -dot- edu

About Me

I am a PhD candidate (2008 - 2013 (expected)) in Computer Science at UC Berkeley. I am a member of the Natural Language Processing group, and my advisor is Dan Klein.

I completed a Master of Science (MS) degree in Computer Science from UC Berkeley. My MS thesis is titled `An All-Fragments Grammar for Simple and Accurate Parsing'. [pdf]

I completed a Bachelor of Technology (BTech) degree in Computer Science and Engineering from the Indian Institute of Technology Kanpur, in May 2008.

Here is a brief CV.

Research Interests

Statistical natural language processing, machine learning, semantics, syntactic parsing, coreference resolution, machine translation, information extraction.

Publications

• Good, Great, Excellent: Global Inference of Semantic Intensities
  Gerard de Melo and Mohit Bansal.
  Accepted for publication in TACL (to be presented at ACL 2013).

• Coreference Semantics from Web Features
  Mohit Bansal and Dan Klein.
  Proceedings of ACL 2012 (Jeju, South Korea).[pdf] [slides]

• Unsupervised Translation Sense Clustering
  Mohit Bansal, John DeNero, and Dekang Lin.
  Proceedings of NAACL 2012 (Montreal, Canada).[pdf] [slides]

• Web-Scale Features for Full-Scale Parsing
  Mohit Bansal and Dan Klein.
  Proceedings of ACL 2011 (Portland, Oregon, USA).[pdf] [slides]

• Gappy Phrasal Alignment By Agreement
  Mohit Bansal, Chris Quirk, and Robert Moore.
  Proceedings of ACL 2011 (Portland, Oregon, USA).[pdf] [poster] [slides]

• The Surprising Variance in Shortest-Derivation Parsing
  Mohit Bansal and Dan Klein.
  Proceedings of ACL 2011 (Portland, Oregon, USA) as a short paper.[pdf] [poster]

• Mention Detection: Heuristics for the OntoNotes annotations
  Jonathan K. Kummerfeld, Mohit Bansal, David Burkett, and Dan Klein.
  Proceedings of CoNLL 2011 (Portland, Oregon, USA) shared task.[pdf]

• Simple, Accurate Parsing with an All-Fragments Grammar
  Mohit Bansal and Dan Klein.
  Proceedings of ACL 2010 (Uppsala, Sweden).[pdf] [slides]

• Efficient Parsing for Transducer Grammars
  John DeNero, Mohit Bansal, Adam Pauls, and Dan Klein.
  Proceedings of NAACL 2009 (Boulder, Colorado).[pdf] [slides]

• The power of negative thinking: Exploiting label disagreement in the min-cut classification framework
  Mohit Bansal, Claire Cardie, and Lillian Lee.
  Proceedings of COLING 2008 (Manchester, UK) as a short paper.[pdf] [slides]

• Estimating Hybrid Frequency Moments of Data Streams
  Sumit Ganguly, Mohit Bansal, and Shruti Dube.
  Proceedings of FAW 2008.
  Also accepted in the Journal of Combinatorial Optimization (JOCO).[pdf]

• Text Processing for Text-to-Speech Systems in Indian Languages:
  Anand A Raj, Tanuja Sarkar, Satish C Pammi, Santhosh Yuvaraj, Mohit Bansal, Kishore Prahallad, and Alan W Black.
  Proceedings of ISCA SSW6 2007.[pdf]

Awards

Outstanding Graduate Student Instructor Award, 2011-2012, UC Berkeley

Qualcomm Innovation Fellowship, 2011

Tong Leong Lim Pre-Doctoral Prize 2011 , EECS, UC Berkeley

INLAKS Fellowship (Award of Excellence at IITs), 2005-2008

OPJEMS Fellowship, IIT Kanpur, 2007-2008

Cornell Summer Research Fellowship, Summer 2007

Academic Excellence Award, IIT Kanpur, 2004-2005 and 2005-2006

Teaching

Graduate Student Instructor for CS188 (Introduction to Artificial Intelligence), Fall 2011.
Instructor: Dan Klein.
Topics covered: Search, CSPs, Games, MDPs, Reinforcement Learning, Probability, Bayes' Nets, VPI, HMMs, Machine Learning

Graduate Student Instructor for CS194-10 (Advanced Topics in Artificial Intelligence), Spring 2009.
Instructors: Pieter Abbeel, Dan Klein, Jitendra Malik.
Topics covered: Computer Vision, Robotics, Natural Language Processing and Machine Learning.

Experience

Google Research, Mountain View, Summer 2011
Mentors: John DeNero and Dekang Lin

Microsoft Research, Redmond, Summer 2010
Mentors: Chris Quirk and Bob Moore

Cornell University (CS), Summer 2007
Mentors: Lillian Lee and Claire Cardie

Contact Information