Negar Arabzadeh
Postdoctoral Fellow at UC Berkeley
This is Negar Arabzadeh. I am a postdoctoral researcher at UC Berkeley Sky Lab, working with Professor Matei Zaharia. I completed my PhD at the University of Waterloo, where I was advised by Dr. Charles Clarke. My research lies at the intersection of information retrieval and the evaluation of—and with—large language models (LLMs) in information access systems.
I am on the academic job market starting in 2026.
News.
2026
- 🌟
“MAP: Measuring Agents in Production” accepted at ICML 2026 as a Spotlight (top 2.2%).
- 🥇
3rd Place, Mathematics Doctoral Prize Competition, University of Waterloo.
- 🏆
Best Paper Award at ECIR 2026 for “ReFormeR: Learning and Applying Explicit Query Reformulation Patterns.”
- 🪑
Reproducibility Track Chair, SIGIR 2026.
- 🏆
Best Industry Talk Award at WSDM 2026 for “Can LLMs Uphold Research Integrity? Evaluating the Role of LLMs in Peer Review Quality.”
- 🪑
Area Chair, NeurIPS 2026.
- 💬
Joined the panel discussion at the WSDM 2026 Doctoral Symposium.
- 🪑
Workshop Chair, ECIR 2026.
- 📚
“QueryGym: A Toolkit for Reproducible LLM-Based Query Reformulation” accepted at The Web Conference 2026.
- 🪑
Publicity Chair, CHIIR 2026.
- 📚
“PeeriScope: A Multi-Faceted Framework for Evaluating Peer Review Quality” accepted at The Web Conference 2026.
- 🪑
Area Chair, ACL Rolling Review (ARR).
- 🗺️
Presented MAP: Measuring Agents in Production at a NICE Talk — insights from a year-long empirical study on real-world agent deployments.
- 🗺️
Presented MAP: Measuring Agents in Production at Google Sense AI.
Right now I am working on agentic IR, RAG-related problems, and using retrieval for reasoning-intensive tasks. Always happy to collaborate.