Scholar
Saayan Mitra
Google Scholar ID: QyslscUAAAAJ
Principal Research Scientist, Adobe
Citations & Impact
Citations
621
H-index
11
i10-index
14
Publications
20
Co-authors
17
list available
Contact
Publications
Resume (English only)
Academic Achievements
- VISIAR: Empower MLLM for Visual Story Ideation (ACL 2025)
- Shape My Moves: Text-Driven Shape-Aware Synthesis of Human Motions (CVPR 2025)
- Token-level adversarial prompt detection based on perplexity measures and contextual information (ICLR Workshop 2025)
- Detecting Ambiguities to Guide Query Rewrite for Robust Conversations in Enterprise AI Assistants (arXiv 2025)
- CodeLutra: Boosting LLM Code Generation via Preference-Guided Refinement (Arxiv 2024)
- TAME-RD: Text Assisted Replication of Image Multi-Adjustments for Reverse Designing (ACL 2024)
- HanDiffuser: Text-to-Image Generation With Realistic Hand Appearances (CVPR 2024)
- A/B testing under Interference with Partial Network Information (AISTATS 2024)
- ToolChain*: Efficient Action Space Navigation in Large Language Models with A* Search (ICLR 2024)
- Decentralized Personalized Online Federated Learning (IEEE Big Data 2023)
- Flash: Concept Drift Adaptation in Federated Learning (ICML 2023)
- Direct Inference of Effect of Treatment (DIET) for a Cookieless World (AISTATS 2023)
- Privacy Aware Experiments without Cookies (WSDM 2023)
Research Experience
- Currently a Principal Research Scientist at Adobe Research, focusing on enhancing AI assistants in Adobe products through large language models. Has worked on multiple projects involving media formats, streaming, and content protection.
Education
- Ph.D. from Iowa State University in 2009.
Background
- Research interests include leveraging large language models to build AI assistants, and how to effectively use contextual and behavioral data to improve content. Previously worked in the domains of recommendations and advertising.
