Scholar

YIRAN WU

Google Scholar ID: 2Gx5IQ4AAAAJ

PhD Student, Pennsylvania State University

Agentic AIReinforcement Learning

Citations & Impact

All-time

Citations

2,126

H-index

i10-index

Publications

Co-authors

list available

Contact

Publications

12 items

Browse publications on Google Scholar (top-right) ↗

Resume (English only)

Academic Achievements

Published 'ExCyTIn-Bench: Evaluating LLM agents on Cyber Threat Investigation'
Published 'SimpleDoc: Multi-Modal Document Understanding with Dual-Cue Page Retrieval and Iterative Refinement'
Published 'Absolute zero: Reinforced self-play reasoning with zero data'
Published 'StateFlow: Enhancing LLM Task-Solving through State-Driven Workflows' (COLM 2024)
Published 'AutoDefense: Multi-Agent LLM Defense against Jailbreak Attacks' (arXiv preprint arXiv:2403.04783)
Published 'AutoGen: Enabling Next-Gen LLM Applications via Multi-Agent Conversation' (COLM 2024)
Published 'MathChat: Converse to Tackle Challenging Math Problems with LLM Agents' (ICLR 2024 Workshop on LLMAgents)
Published 'Unified off-policy learning to rank: a reinforcement learning perspective' (NeurIPS 2023)
Published 'Automated object detection in experimental data using combination of unsupervised and supervised methods' (Frontiers in Physiology)
Led or contributed to multiple high-star open-source projects including AutoGen (50k stars) and Absolute Zero (1.7k stars)

Research Experience

Research Intern at Microsoft Research, Redmond (2024)
Research Intern at Microsoft Research, Redmond (2025)
Co-creator and maintainer of open-source LLM agents framework AutoGen (now AG2)
Co-creator and maintainer of Absolute Zero RLVR training for LLMs
Creator of ExCyTIn-Bench: the first benchmark for evaluating LLM agents on cyber threat investigation tasks

Co-authors

4 total