Hi, I am Pritika. I’m currently working as a Research Associate at Adobe Research. I graduated with a Bachelor of Engineering in Computer Science from Birla Institute of Technology and Science, Pilani (BITS, Pilani). I am interested in Human-LM interaction for creative tasks.
Research
At Adobe Research, I primarily work with Dr. Aparna Garimella and Dr. Balaji V. Srinivasan on document experience projects which involve creation and transformation of documents into other consumable formats (reports, slides, infographics). My focus areas include ambiguity resolution, personalization, factuality, content planning and generation of structured content (tables and charts).
I completed my bachelor’s thesis under the supervision of Prof. Lifu Huang at Virginia Tech. My project focused on the extraction of key-value pairs from domain and language agnostic visually rich documents, utilizing NLP and graph techniques. During my undergraduate studies, I explored various research fields including biometric authentication, route optimization, and time-series forecasting under the guidance of Prof. Kamlesh Tiwari.
Publications and Preprints
Is This a Bad Table? A Closer Look at the Evaluation of Table Generation from Text
EMNLP 2024 (short)
Pritika Ramu, Aparna Garimella, Sambaran Bandyopadhyay
Enhancing Post-Hoc Attributions in Long Document Comprehension via Coarse Grained Answer Decomposition
EMNLP 2024
Pritika Ramu, Koustava Goswami, Apoorv Saxena, Balaji Vasan Srinivasan
Unraveling the Truth: Do LLMs really Understand Charts? A Deep Dive into Consistency and Robustness
EMNLP 2024 Findings
Srija Mukhopadhyay, Adnan Qidwai, Aparna Garimella, Pritika Ramu, Vivek Gupta, Dan Roth
Zooming in on Zero-Shot Intent-Guided and Grounded Document Generation using LLMs
INLG 2024 (oral presentaion)
Pritika Ramu, Pranshu Gaur, Rishita Emandi, Himanshu Maheshwari, Danish Javed and Aparna Garimella
RE2: Region-Aware Relation Extraction from Visually Rich Documents
NAACL 2024
Pritika Ramu, Sijia Wang, Lalla Mouatadid, Joy Rimchala, Lifu Huang
Patents
Planning-based Method to Assist Intent-guided and Grounded Document Generation using LLMs
Pritika Ramu, Aparna Garimella, Himanshu Maheshwari
US Patent Application 18/891,558; Adobe Inc.
Minimally Guided Semantic Extraction
Inderjeet Nair, Pritika Ramu, Akshay Singhal, Kumud Lakara, Vikas Balani, Anandhavelu N
US Patent Application 18/360,721; Adobe Inc.
Multimodal Grounded Content Expansion
Pranshu Gaur, Danish Javed, Rishita Emandi, Himanshu Maheshwari, Pritika Ramu, Aparna
Garimella
US Patent Application to be filed; Adobe Inc.
CV
Work Experience
Adobe Research, Bangalore
Research Associate - Jul 2023 - Present
VT-NLP Lab, Blacksburg
Research Assistant - Jan 2023 - Jun 2023
Adobe Research, Bangalore
Research Intern — Jun 2022 - August 2022
Goibibo, Bangalore
Technology Development Intern — Jun 2021 - Aug 2021