I am a PhD student in Computer Science at Arizona State University, working with Dr. Vivek Gupta in the Complex Data Reasoning and Analysis (CoRAL) Lab.

My research focuses on personalized, explainable, and trustworthy evaluation for multimodal large language models (MLLMs). I study how agent-driven evaluation frameworks can adapt metrics to user intent and task context, while uncovering risks such as multimodal bias, visual misinterpretation, and cross-document reasoning failures.

Previously, I was a Research Scientist Intern at Adobe Research, where I worked on LLM-agent pipelines for holistic multimodal and multidocument understanding.

• Multimodal LLM Evaluation
• Agentic Reasoning & LLM Agents
• Explainable & Trustworthy AI
• Structured and Open-World Reasoning
• Computer Vision & 3D Understanding
• Representation & Continual Learning

I have published 10+ peer-reviewed papers at venues including
ACL, AACL, WACV, CVPR Workshops, ICCV Workshops, SIGGRAPH Asia, AAAI.

👉 See full list on my Google Scholar or website.

• Agent-driven, referenceless evaluation frameworks for multimodal LLMs
• Intent-aware explainability for tabular and chart-based reasoning
• Robustness & grounding analysis for multimodal agents

Languages: Python, C++, C, Bash
ML / AI: PyTorch, HuggingFace, vLLM, OpenAI, scikit-learn
Vision / Graphics: OpenCV, Blender
Research: LaTeX, Git, Linux, Jupyter, Obsidian

• 🌍 Website: https://tejasanvekar.github.io
• 💼 LinkedIn: https://www.linkedin.com/in/tejas-anvekar-b88b461a4
• 🐦 Twitter/X: https://twitter.com/anvekar_tejas
• 📫 Email: tanvekar@asu.edu