Negar Arabzadeh
Postdoctoral Fellow at UC Berkeley
This is Negar Arabzadeh. I am a postdoctoral researcher at UC Berkeley, 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.
January 2025
📚 2 papers get accepted at The Web Conf2026:
-
“QueryGym: A Toolkit for Reproducible LLM-Based Query Reformulation”
-
W/ Amin Bigdeli, Radin Hamidi Rad, Mert Incesu, Charles L. A. Clarke, and Ebrahim Bagheri
-
-
“PeeriScope: A Multi-Faceted Framework for Evaluating Peer Review Quality”
-
W/ Sajad Ebrahimi, Soroush Sadeghian, Ali Ghorbanpour, Sara Salamat, Seyed Mohammad Hosseini, Hai Son Le, Mahdi Bashari, and Ebrahim Bagheri
-
🗺️ We presented MAP: Measuring Agents in Production at a NICE Talk, sharing insights from a year-long empirical study on real-world agent deployments.
🗺️ We presented MAP: Measuring Agents in Production at Google Sense AI.
December 2025
🏅Received Runner-up Best paper award for SIGIR AP 2026 for "Adversarial Attacks against Neural Ranking Models via In-Context Learning"
✈️ I attended my first Neurips! It was pretty OVERWHELMING.
🪧 Presented three workshop papers At Neurips workshops:
- Restructuring the Corpus Makes RAG Work for Math at MATH AI workshop
-DeepScholar-Bench: A Live Benchmark and Automated Evaluation for Generative Research Synthesis at LLM Evaluation Workshop
- Towards Automatically Optimizing Retrieval Augmented AI Systems at ML for systems workshop
🗺️ We released MAP: Measuring agents in Production! This was a whole year long project!
📚Our Workshop proposal got accepted at NeurIPS 2025:
“Vision Language Models: Challenges of Real-World Deployment”
W/ Mozhgan Nasr Azadani, Yimu Wang, Krzysztof Czarnecki, Richard He Bai, and Lukas Schmid
📚Our paper “Can LLMs Uphold Research Integrity? Evaluating the Role of LLMs in Peer Review Quality” is under final proceedings preparation for WSDM 2026.
W/ Negar Arabzadeh, Sajad Ebrahimi, Soroush Sadeghian, Seyed Mohammad Hosseini, Alireza Daqiq, Hai Son Le, Mahdi Bashari, and Ebrahim Bagheri
November 2025
🏋️♀️ We released "QueryGym: A Toolkit for Reproducible LLM-Based Query Reformulation" w/ Amin Bigdeli, Radin Hamidi Rad, Mert Incesu, Charles LA Clarke and Ebrahim Bagheri
✈️ Attended CIKM in Seoul, South Korea, where I presented one applied research paper, one short paper, and delivered an industry talk.
October 2025
🎓 Graduated with my PhD from the University of Waterloo, with a nomination for the Gold Thesis Award.
September 2025
🪑I will be serving as Reproducibility track chair for SIGIR 2026.
📚 Our paper "Query Performance Prediction Using Neural Query Space Proximity" got accepted at TIST jounal.
W/ Amin Bigdeli, Sajad Ebrahimi, Sara Salamat, Shirin SeyedSalehi, Maryam Khodabakhsh, Fattane Zarrinkalam, Ebrahim Bagheri
August 2025
🌟 First project release as a postdoc at Sky Lab, a Stanford + Berkeley collaboration! DeepScholar-Bench, a live benchmark for generative research synthesis.
🪑I will be serving as Workshop chair for ECIR 2026.
📚 3 Papers get accepted at CIKM 2025:
-
RottenReviews: Benchmarking Review Quality with Human and LLM-Based Judgments
-
w/ Sajad Ebrahimi, Soroush Sadeghian, Ali Ghorbanpour, Sara Salamat, Muhan Li, Hai Son Le, Mahdi Bashari and Ebrahim Bagheri
-
-
Building Trustworthy Peer Review Quality Assessment Systems
-
w/ Sajad Ebrahimi, Ali Ghorbanpour, Soroush Sadeghian, Sara Salamat, Muhan Li, Hai Son Le, Mahdi Bashari and Ebrahim Bagheri
-
-
LLM-as-a-Judge in Entity Retrieval: Assessing Explicit and Implicit Relevance
-
w/ Mohammad Hossein Saliminabi, Seyed Mohammad Hosseini, Dimitrios Androutsos, Morteza Zihayat and Ebrahim Bagheri
-
July 2025
✈️I attended SIGIR in Padua,Italy.
🏅I won best reviewer award at SIGIR.
June 2025
🪑I will be serving as Publicity chair for CHIIR 2026.
🏆 I was honored to receive SIGIR student travel grant.