Scholar
Amith Ananthram
Google Scholar ID: _dRPJYkAAAAJ
Columbia University
Citations & Impact
Citations
107
H-index
6
i10-index
3
Publications
13
Co-authors
0
Publications
Resume
Academic Achievements
- Selected Publications: 1. PoSh: Using Scene Graphs to Guide LLMs-as-a-Judge for Detailed Image Descriptions; 2. Mining Contextualized Visual Associations from Images for Creativity Understanding; 3. See It from My Perspective: How Language Affects Cultural Bias in Image Understanding; 4. Data Caricatures: On the Representation of African American Language in Pretraining Corpora; 5. Enhancing Multimodal Affective Analysis with Learned Live Comment Features; 6. FeelingBlue: a Corpus for Understanding the Emotional Connotation of Color in Context.
Research Experience
- Industry Experience: Full-stack software engineer at Stripe and Wealthfront, led cross-functional projects that delivered thoughtful experiences with rigorous technical solutions. Research Projects: Developed PoSh, an interpretable & replicable metric for detailed image descriptions; Introduced DOCENT, a new dataset of artwork with expert descriptions and judgments from art history students; Collaborating with a team at the National Gallery of Art to expand accessibility in their collection.
Education
- Degree: PhD; School: Columbia University; Advisor: Professor Kathleen McKeown; Time: Not specified; Major: Computer Science.
Background
- Research Interests: Architectures, pre/post-training and evaluation methods for vision-language models; Language and its role in vision. Background: PhD candidate in Computer Science at Columbia University, advised by Professor Kathleen McKeown. Research focuses on vision-language models, particularly the strengths and limitations of different approaches to multimodal alignment. Recent focus has been on detailed image description with an emphasis on works of art.
Miscellany
- Personal Interests: Enjoys building reliable, maintainable systems that drive value for end users.