Photo of Emre Can Acikgoz

Emre Can Acikgoz

I like scaling agents on large environments.

PhD Student at UIUC NLP
Advised by Gokhan Tur and Dilek Hakkani-Tür

Research Intern at Microsoft Research
Advised by Baolin Peng and Jianfeng Gao

Email Scholar GitHub Hugging Face Twitter LinkedIn

Emre Can Acikgoz is a PhD student in Computer Science at UIUC, advised by Gokhan Tur and Dilek Hakkani-Tür. He is currently a research intern in the Deep Learning Group at MSR, supervised by Baolin Peng and Jianfeng Gao. His research focuses on self-evolving LLM agents and agentic learning, with broader interests in post-training dynamics, self-play, and test-time learning. His work has been published at top-tier venues including but not limited to NeurIPS, ICLR, ACL, and EMNLP; representative works include ToolRL, CoALM, TT-SI, and Tool-R0. He co-organizes Lifelong Agents: Learning, Aligning, and Evolving at ICLR 2026 and COLM 2026. He is co-chair of the Student Research Workshop (SRW) at ACL 2027. Other than LLM agents, he also likes scaling the limits of a human body: he is an active crossfitter, snowboarder, and a former college athlete.

"He begins working calculus problems in his head as soon as he awakens. He did calculus while driving in his car, while sitting in the living room, and while lying in bed at night." — from the divorce complaint of Richard Feynman's second wife

Project Lifelong Agent

The broader vision shaping my PhD is agents that can continually learn, quickly adapt, and generalize with minimal supervision. Lifelong Agents project has three pillars: Learning, Aligning, and Evolving. I focus primarily on the Evolving pillar.

Project page

News

Publications /

* denotes equal contribution. Full list on Google Scholar.

Industry experience

Service

Organizing

Reviewing

Teaching