Scholar
Emre Can Acikgoz
Google Scholar ID: l6h_3H8AAAAJ
University of Illinois Urbana-Champaign
Citations & Impact
Citations
152
H-index
7
i10-index
4
Publications
12
Co-authors
3
list available
Publications
Resume (English only)
Academic Achievements
- - 2025, arXiv, 'Self-Improving LLM Agents at Test-Time', Authors: Emre Can Acikgoz* et al.
- - 2025, SIGDial, 'TD-EVAL: Revisiting Task-Oriented Dialogue Evaluation by Combining Turn-Level Precision with Dialogue-Level Comparisons', Authors: Emre Can Acikgoz* et al.
- - 2025, NeurIPS, 'ToolRL: Reward is All Tool Learning Needs', Authors: Cheng Qian*, Emre Can Acikgoz* et al.
- - 2025, arXiv, 'A Desideratum for Conversational Agents: Capabilities, Challenges, and Future Directions', Authors: Emre Can Acikgoz* et al.
- - 2025, ACL, 'SMART: Self-Aware Agent for Tool Overuse Mitigation', Authors: Cheng Qian*, Emre Can Acikgoz* et al.
Research Experience
- - Currently an Applied Scientist Intern at Amazon Alexa AI (Bellevue)
- - Focused on research of Self-Evolving Agents during his PhD
- - Conducted research on Large Language Models and Post-Training for his MSc
- - Participated in research on Supervised and Unsupervised Morphological Analysis during his BSc
Education
- - PhD, University of Illinois at Urbana-Champaign (UIUC), Advisors: Dilek Hakkani-Tur and Gokhan Tur
- - MSc, Koc University, Major: Computer Science, Advisor: Deniz Yuret, Focus: Large Language Models and Post-Training
- - BSc, Koc University, Major: Electrical and Electronics Engineering (AI focus), Advisor: Deniz Yuret, Focus: Supervised and Unsupervised Morphological Analysis
Background
- Research Interests: Self-Evolving Agents, Agent Learning, and Test-Time Training. Field: Computer Science, focusing on Large Language Models and Post-Training.
Miscellany
- Feel free to schedule a short chat-meeting if you want to discuss your research or seek advice/feedback (non-industrial).
