|
I am an MS Student at UC San Diego and a Jr. Applied Scientist at Amazon, working on Reasoning, Agents, and ML systems. Before I joined UCSD, I spent two amazing years as a Data/Applied Scientist building Search Recommendation Systems (0 to 1) at Meesho in India. I have been instrumental in developing query understanding systems (search query correction, autocomplete), multimodal information retrieval models, and ranking systems for e-commerce. I have expertise in developing & deploying large-scale models and data pipelines serving 130M+ monthly active users. Working on ML models serving millions of users has shown me firsthand the importance of developing accurate models and ensuring they are efficient and scalable in real-world applications. I graduated from the Indian Institute of Technology, Guwahati, in 2022 with a major in Electronics and Communication Engineering and a minor in Mathematics. I also pursued research on investigating why language models struggle with temporal and numerical reasoning on semi-structured data. |
Publications
AMA-Bench: Evaluating Long-Horizon Memory for Agentic Applications
Yujie Zhao, Boqin Yuan, Junbo Huang, Haocheng Yuan, Zhongming Yu, Haozhou Xu, Lanxiang Hu, Abhilash Shankarampeta, Zimeng Huang, Wentao Ni, Yuandong Tian, Jishen Zhao.
Under review, ICML
Benchmarking Scientific Understanding and Reasoning for Video Generation using VideoScience-Bench
Lanxiang Hu, Abhilash Shankarampeta, Yixin Huang, Zilin Dai, Yujie Zhao, Haoqiang Kang, Haoyang Yu, Daniel Zhao, Tajana Rosing, Hao Zhang.
Under review, CVPR
[paper]
Evidence-Guided Schema Normalization for Temporal Tabular Reasoning
Ashish Thanga, Vibhu Dixit, Abhilash Shankarampeta, Vivek Gupta.
Under review, ACL
[paper]
Towards Interpretable and Inference-Optimal COT Reasoning with Sparse Autoencoder-Guided Generation
Daniel Zhao*, Abhilash Shankarampeta*, Lanxiang Hu*, Tajana Rosing, Hao Zhang.
Preprint
[paper]
TRANSIENTTABLES: Evaluating LLMs’ Reasoning on Temporally Evolving Semi-structured Tables
Abhilash Shankarampeta*, Harsh Mahajan*, Tushar Kataria, Dan Roth, Vivek Gupta.
Nations of the Americas Chapter of the Association for Computational Linguistics (NAACL), April 2025.
[homepage] [paper] [code]
Exploring the numerical reasoning capabilities of language models: A comprehensive analysis on tabular data
Mubashara Akhtar*, Abhilash Shankarampeta*, Vivek Gupta, Arpit Patil, Oana Cocarascu, Elena Simperl.
Empirical Methods in Natural Language Processing (EMNLP), December 2023.
[paper] [code]
Enhancing Tabular Reasoning with Pattern Exploiting Training
Abhilash Shankarampeta, Vivek Gupta, Shuo Zhang.
Asia-Pacific Chapter of the Association for Computational Linguistics (AACL), November 2022.
SUKI: Structured and Unstructured Knowledge Integration Workshop at NAACL, July 2022 (non-archival).
[homepage] [paper] [slides] [poster] [media]
Few-Shot Class Incremental Learning with Generative Feature Replay
Abhilash Shankarampeta, Koichiro Yamauchi.
International Conference on Pattern Recognition Applications and Methods (ICPRAM), February 2021.
[paper] [code]
Blog Posts
Embedding Based Retrieval for Semantic Search at Meesho
Mahendra Singh Meena and Abhilash Shankarampeta, December, 2023.