S I Harini

I am a software engineer at Adobe. Previously, I was an intern at Adobe MDSR where I worked at the intersection of human behavior and VLMs with my collaborators Somesh and Yaman . Currently, I am interested in improving System 2 capabilities of LLMs/VLMs, specifically for tasks that involve complex-reasoning, planning and decision-making.

Previously, I obtained my bachelor's in CS from BITS Pilani, Goa. For my undergraduate thesis, I worked on "text-to-simulation-code" generation with Dr. Nick Haber at Stanford. At BITS, I was mentored by Dr. Ashwin Srinivasan and Dr. Gautam Shroff, where I explored neurosymbolic techniques to solve the ARC challenge.

Outside of research, I enjoy reading and travelling and I also love playing badminton.

Google Scholar  /  GitHub  /  Resume  /  Twitter  /  Linkedin

News
  • 2025-01 Behavior LLava and Transsuasion accepted to ICLR'2025!
  • 2024-10 Long-Term Ad Memorability: Understanding & Generating Memorable Ads, is accepted to WACV'2025.
  • 2024-10 Released PersuasionArena A large-scale automated leaderboard of LLM's persuasion capabilities.
  • 2024-10 FactorSim, is accepted to NeurIPS'2024.
  • 2024-07 Graduated from BITS Pilani with a Bachelor's degree in Computer Science.
  • 2024-05 Released BehaviorLLaVA, training VLMs on human behavior improves performance on close to 50 tasks.
  • 2023-05 Joined Adobe as a research intern, working on multi-modal foundation models