I am a Machine Learning Engineer at Amazon Alexa where I work on developing content moderation models for Alexa Large Language Models (LLM) as well as research on mitigating security vulnerabilities (such as prompt injection attacks) on large action models (LAMs). I completed my MS in Information Technology from The Polytechnic School, Arizona State University. I was a research volunteer working with Prof. Chitta Baral and Prof. Paulo Shakarian. I closely collaborate with Dr. Swaroop Mishra and Dr. Arindam Mitra. My research interests are Efficient Large Language Models (LLMs), LLM Alignment, Augment LLM Reasoning and RAGs. I am also exploring vision language research. Please find my resume attached here

Research

My research interests include LLM Reasoning, LLM Security, Instruction Tuning, Creation & Curation of Benchmarks, Preference Optimization and Model Efficiency. My strength lies in generating new ideas, and I am fortunate to collaborate with a diverse set of awesome researchers.

News

• 06.2024 Started Full Time Job as Machine Learning Engineer at Amazon Alexa
• 02.2024 Started Full Time Job as Software Developer at Amazon Web Services
• 05.2023 Started Internship at Amazon Web Services

Papers

• PolyMATH: A Challenging Multi-modal Mathematical Reasoning Benchmark
  Himanshu Gupta*, Shreyas Verma*, Ujjwala Anantheswaran*, Kevin Scaria*, Mihir Parmar, Saurabh Arjun Sawant, Swaroop Mishra, Chitta Baral
  Submitted to ICLR 2025 (* - Equal First Author Contribution)
  

• Investigating the Robustness of LLMs on Math Word Problems
  Ujjwala Anantheswaram, Kevin Scaria, Himanshu Gupta, Swaroop Mishra, Chitta Baral
  preprint
  

• Embarrassingly Simple Unsupervised Aspect Based Sentiment Tuple Extraction
  Kevin Scaria, Abyn Scaria, Ben Scaria
  preprint
  

• TarGEN: Targeted Data Generation with Large Language Models
  Himanshu Gupta, Kevin Scaria, Ujjwala Anantheswaran, Shreyas Verma, Mihir Parmar, Saurabh Arjun Sawant, Swaroop Mishra, Chitta Baral
  COLM 2024
  

• Rule-Based Error Detection and Correction to Operationalize Movement Trajectory Classification
  Bowen Xi, Kevin Scaria, Paulo Shakarian
  Submitted to ICML 2024
  

• InstructABSA: Instruction Learning for Aspect Based Sentiment Analysis
  Kevin Scaria, Himanshu Gupta, Siddharth Goyal, Saurabh Arjun Sawant, Swaroop Mishra, Chitta Baral
  NAACL 2024
  

• “John is 50 years old, can his son be 65?” Evaluating NLP Models’ Understanding of Feasibility.
  Himanshu Gupta, Neeraj Varshney, Swaroop Mishra, Kuntal Kumar Pal, Saurabh Arjun Sawant, Kevin Scaria, Siddharth Goyal, Chitta Baral
  EACL 2023
  

• Please check my google scholar page for all the papers.

Education

• 01.2022-12.2023 MS - Information Technology, Arizona State University
• 2019-2019 Post Graduate Program in Data Science Engineering, Great Lakes Institute of Management
• 2014-2018 B.Tech., SRM Institute of Science and Technology,

Experience

• 06.20234-Present Machine Learning Engineer at Amazon Alexa.
• 02.20234-Present Software Engineer at Amazon Web Services.
• 05.2023-07.2023 Software Engineer Intership at Amazon Web Services.
• 03.2022-12.2023 Data Science Analyst - ASU Decision Theater Network.
• 08.2022-04.2023 Graduate research assistant under Dr. Paulo Shakarian - ASU Lab V2.
• 01.2022-04.2023 Graduate research volunteer under Dr. Chitta Baral ASU CogInt Labs.
• 01.2020-12.2021 Data Science Analyst at an AI consulting firm - Tiger Analytics.

Volunteering, Honors and Awards

• 2024 NeurIPS RBFM Workshop Reviewer
• 2023 EMNLP Reviewer