Rajaswa Patil

Research Fellow

Microsoft PROSE

About Me

I am a Research Fellow at the Microsoft PROSE team. Within PROSE, I a work with the Blue-Pencil team, which focuses on developing assistive modeless program synthesis systems. My research & development interests lie in Artificial Intelligence, Computational Linguistics, and Formal Languages. I have completed my bachelors degree in Electrical & Electronics engineering from BITS Pilani (Goa Campus), India. Besides my professional research interests, I also love to spend time reading books, hitting the gym, and participating in education efforts for technology. I am also occasionally curious about behavioral science, economics, and anthropology.

Interests

Natural Language Processing
Neuro-symbolic AI
Cognitive Science

Education

B.E. in Electrical & Electronics Engineering, 2021

BITS Pilani, K. K. Birla Goa Campus

News

I will be serving as a Mentor at the NAACL 2022 Student Research Workshop!

I will be serving as a Mentor at the NAACL 2022 Student Research Workshop. If you are currently a student, consider submitting your completed work, or work in progress to the workshop.

Jan 15, 2022

I will be serving as a Mentor at the ACL 2022 Student Research Workshop!

I will be serving as a Mentor at the ACL 2022 Student Research Workshop. If you are currently a student, consider submitting your completed work, or work in progress to the workshop.

Dec 20, 2021

Long Paper accepted at MRL-EMNLP 2021!

Our paper Vyākarana: A Colorless Green Benchmark for Syntactic Evaluation in Indic Languages has been accpeted for presentation at the First Workshop on Multilingual Representation Learning at EMNLP 2021. You can check out the code and dataset here!

Sep 25, 2021

I'll be serving as a PC member at the MRL 2021 workshop at EMNLP 2021

I will be serving as the member of the Program Committe for the First Workshop on Multilingual Representation Learning at EMNLP 2021.

Aug 25, 2021

I'll be serving as a PC member at the MRL 2021 workshop at EMNLP 2021

See all

Experience

Research Fellow

Microsoft PROSE

Oct 2022 – Present Remote

Working on code-suggestions with LLMs in collaboration with the Visual Studio and .NET teams.

Pre-doctoral Research Fellow

TCS Research

Jul 2021 – Sep 2022 Pune, IN

Worked on broad research topics: NLP for Programming, Text-to-SQL Semantic Parsing, and Chart Question-Answering.

Data Scientist (Intern)

Cognizer Inc.

Apr 2021 – Jun 2021 Bengaluru, IN

Developed APIs & data resources for Question-generation, Grammar-correction, and Text-simplification for the Cognizer Genius platform.

Research Intern

Max Planck Institute for Human Cognitive and Brain Sciences

Apr 2021 – Jun 2021 Leipzig, DE (Remote)

Worked on Combinatory Categorial Grammar (CCG) parsing.

Research Assistant

MIDAS Lab - IIITD

Jun 2020 – Jun 2021 New Delhi, IN

Worked on modelling various aspects of ESL speakers' langauge Ex: language complexity, discourse coherence, language fluency, etc. under a collaboration with [SLTI Inc.](https://www.linkedin.com/company/2lti/)

Research Intern

MIDAS Lab - IIITD

Dec 2019 – May 2020 New Delhi, IN

Worked on building neural network based speech proficiency scoring models for ESL speakers.

Project Assistant

Cognitive Neuroscience Lab - BITS Goa

Sep 2019 – May 2021 Goa, IN

Lead various projects on Syntactic Processing for Indic languages, Grounded Language Learning, Mental Lexicon Development and EEG-based brain imagery.

Summer Research Intern

IIT Bombay

May 2019 – Jul 2020 Mumbai, IN

Worked on a research project under the NCETIS and the Dept. of Computer Science & Engineering, IIT Bombay. Worked on modelling aspect-based sentiment analysis on open-domain web data. Developed an end-to-end pipeline for data-domain independent opinion-mining for natural language texts.

Recent Publications

Quickly discover relevant content by filtering publications.

Program Synthesis for Complex QA on Charts via Probabilistic Grammar Based Filtered Iterative Back-Translation

Answering complex reasoning questions from chart images is a challenging problem requiring a combination of natural language …

Shabbirhussain Bhaisaheb, Shubham Paliwal, Rajaswa Patil, Manasi Patwardhan, Lovekesh Vig, Gautam Shroff

Program Synthesis for Complex QA on Charts via Probabilistic Grammar Based Filtered Iterative Back-Translation

Probing Semantic Grounding in Language Models of Code with Representational Similarity Analysis

Representational Similarity Analysis is a method from cognitive neuroscience, which helps in comparing representations from two …

Shounak Naik, Rajaswa Patil, Swati Agarwal, Veeky Baths

Probing Semantic Grounding in Language Models of Code with Representational Similarity Analysis

Exploring Dimensions of Generalizability and Few-shot Transfer for Text-to-SQL Semantic Parsing

Existing work on generalization in Text-to-SQL semantic parsing has been restricted to a zero-shot cross-domain setting. In this paper, …

Rajaswa Patil, Manasi Patwardhan, Shirish Karande, Lovekesh Vig, Gautam Shroff

Exploring Dimensions of Generalizability and Few-shot Transfer for Text-to-SQL Semantic Parsing

DRIFT: A Toolkit for Diachronic Analysis of Scientific Literature

In this work, we present to the NLP community, and to the wider research community as a whole, an application for the diachronic …

Abheesht Sharma, Gunjan Chhablani, Harshit Pandey, Rajaswa Patil

DRIFT: A Toolkit for Diachronic Analysis of Scientific Literature

Using Diachronic Distributed Word Representations as Models of Lexical Development in Children

Recent work has shown that distributed word representations can encode abstract semantic and syntactic information from child-directed …

Arijit Gupta, Rajaswa Patil, Veeky Baths

Using Diachronic Distributed Word Representations as Models of Lexical Development in Children

See all publications

Projects

Explore all the projects here.

DRIFT

DRIFT is a tool for Diachronic Analysis of Scientific Literature. The application offers user-friendly and customizable utilities for two modes - Training and Analysis.

Feedback and Memory in Transformers

My final project submission for the Meta Learning course at BITS Goa (conducted by TCS Research & BITS Goa). The project is based on the Feedback Transformer paper.