SHAYAN TALAEI

I am a PhD student at Stanford, very fortunate to be advised by Amin Saberi and Azalia Mirhoseini.

My recent research has focused on building more intelligent and efficient ML systems through multi-agent designs, scaling test-time computation, and bringing these capabilities to real-world applications. I have worked on Text-to-SQL systems, authoring CHESS, CHASE-SQL, and Reasoning-SQL with high efficiency and state-of-the-art accuracy on the BIRD leaderboard.

I received my Bachelor of Science in Computer Engineering from Sharif University of Technology. During my undergraduate studies, I had the opportunity to intern at IST Austria working on distributed optimization and federated learning under the supervision of Dan Alistarh, and at EPFL Switzerland on theoretical machine learning under the supervision of Emmanuel Abbe.

Outside of research, I enjoy reading books, playing table tennis, and sometimes solving and designing Mathematical Olympiad problems.

Research

* denotes equal contribution (Show all papers) (Show selected papers)

SPRINT: Enabling Interleaved Planning and Parallelized Execution in Reasoning Models
Emil Biju*, Shayan Talaei*, Zhemin Huang*, Mohammadreza Pourreza, Azalia Mirhoseini, Amin Saberi (2025)
COLM 2025, SCALR workshop

StorySage: Conversational Autobiography Writing Powered by a Multi-Agent Framework
Shayan Talaei, Meijin Li, Kanu Grover, James Kent Hippler, Diyi Yang, Amin Saberi (2025)
UIST 2025. (Code)

When to Trust Context: Self-Reflective Debates for Context Reliability
Zeqi Zhou*, Fang Wu*, Shayan Talaei*, Haokai Zhao, Cheng Meixin, Tinson Xu, Amin Saberi, Yejin Choi (2025)

Reasoning-SQL: Reinforcement Learning with SQL Tailored Partial Rewards for Reasoning-Enhanced Text-to-SQL
Mohammadreza Pourreza*, Shayan Talaei*, Ruoxi Sun, Xingchen Wan, Hailong Li, Azalia Mirhoseini, Amin Saberi, Sercan O. Arik (2025)
COLM 2025

CHASE-SQL: Multi-Path Reasoning and Preference Optimized Candidate Selection in Text-to-SQL
Mohammadreza Pourreza*, Hailong Li*, Ruoxi Sun, Yeounoh Chung, Shayan Talaei, Gaurav Tarlok Kakkar, Yu Gan, Amin Saberi, Fatma Özcan, Sercan Ö. Arık (2024)
ICLR 2025

CHESS: Contextual Harnessing for Efficient SQL Synthesis
Shayan Talaei, Mohammadreza Pourreza, Yu-Chen Chang, Azalia Mirhoseini, Amin Saberi (2024)
ICML 2025, MAS workshop. (Code)

OptiMUS-0.3: Using Large Language Models to Model and Solve Optimization Problems at Scale
Ali AhmadiTeshnizi, Wenzhi Gao, Herman Brunborg, Shayan Talaei, Connor Lawless, Madeleine Udell (2024)

Hybrid Decentralized Optimization: First- and Zeroth-Order Optimizers Can Be Jointly Leveraged for Faster Convergence
Shayan Talaei*, Matin Ansaripour*, Giorgi Nadiradze, Dan Alistarh (2022)
AAAI 2024. (Code)

Communication-Efficient Federated Learning With Data and Client Heterogeneity
Hossein Zakerinia, Shayan Talaei, Giorgi Nadiradze, Dan Alistarh (2023)
AISTATS 2024. (Code)

Awards