Jiin Woo

Email: jiinw [at] andrew [dot] cmu [dot] edu
LinkedIn / Google Scholar

---

I am an Applied Scientist at Neuron Science team in Amazon Annapurna Labs. I received my PhD in Electrical and Computer Engineering department at Carnegie Mellon University (CMU), fortunate to be advised by Yuejie Chi and Gauri Joshi. Before CMU, I was a machine learning engineer at NAVER Search. Previously, I received my MS in Electrical Engineering (advised by Yung Yi) and BS in Mathematical Sciences at Korea Advanced Institute of Science and Technology (KAIST).

My research interests lie in LLM post-training and collaborative reinforcement learning.

---

Selected Publications/Preprints

* : equal distribution

• Exploration via Exploitation: The Blessing of Reward Diversity in Personalized Federated RL
  Jiin Woo, Gauri Joshi, Yuejie Chi
  NeurIPS, 2026

• Federate the Router: Learning Language Model Routers with Sparse and Decentralized Evaluations
  Baris Askin, Shivam Patel, Anupam Nayak, Andrea Vigano, Jiin Woo, Gauri Joshi, Carlee Joe-Wong
  NeurIPS, 2026
  [arxiv]

• TritonRL: Training LLMs to Think and Code Triton Without Cheating
  Jiin Woo, Shaowei Zhu, Allen Nie, Zhen Jia, Yida Wang, Youngsuk Park
  COLM, 2026
  [paper][arxiv]

• Sample Complexity of Average-Reward Q-Learning: From Single-agent to Federated Reinforcement Learning
  Yuchen Jiao^*, Jiin Woo^*, Gen Li, Gauri Joshi, Yuejie Chi
  [arxiv], 2026 \

• Large Language Model-Enhanced RL for Diverse and Novel Recommendations
  Jiin Woo, Alireza Bagheri Garakani, Tianchen Zhou, Zhishen Huang, Yan Gao
  ARLET Workshop @ NeurIPS, 2025
  [arxiv]

• The Blessing of Heterogeneity in Federated Q-Learning: Linear Speedup and Beyond
  Jiin Woo, Gauri Joshi, Yuejie Chi
  JMLR, 2025
  [paper][arxiv]

• Federated Offline Reinforcement Learning: Collaborative Single-Policy Coverage Suffices
  Jiin Woo, Laixi Shi, Gauri Joshi, Yuejie Chi
  ICML, 2024
  [paper][arxiv]

• The Blessing of Heterogeneity in Federated Q-Learning: Linear Speedup and Beyond
  Jiin Woo, Gauri Joshi, Yuejie Chi
  ICML, 2023
  [paper][slides]

• Iterative Learning of Graph Connectivity from Partially‑Observed Cascade Samples
  Jiin Woo, Jungseul Ok, Yung Yi
  ACM MobiHoc, 2020
  [paper]

• Information Source Finding in Networks: Querying with Budgets
  Jaeyoung Choi, Sangwoo Moon, Jiin Woo, KyungHwan Son, Jinwoo Shin, Yung Yi
  IEEE/ACM Transactions on Networking, 2020
  [paper]

• Rumor Source Detection under Querying with Untruthful Answers
  Jaeyoung Choi, Sangwoo Moon, Jiin Woo, KyungHwan Son, Jinwoo Shin, Yung Yi
  IEEE INFOCOM, 2017
  [paper]

---

Work Experience

• AWS (Annapurna Labs)
  Applied Scientist Intern, Santa Clara, CA, USA, May 2025 - present
  • LLM post-training for kernel code generation
• Amazon (Central Machine Learning)
  Applied Scientist Intern, Seattle, WA, USA, May 2024 - Aug 2024
  • LLM-enhanced RL for recommendation systems
• NAVER Corp. (Search)
  Machine Learning Engineer, Seongnam, Korea, Sep 2018 - Aug 2021
  • Long-tail search query analysis
  • User search behavior/intent analysis
  • Personalized keyword recommendation

---

Awards & Scholarships

• Hsu Chang Memorial Fellowship (2022-2023)
• Carnegie Institute of Technology Dean’s Fellowship (2021-2022)
• KAIST Support Scholarship (2016-2018)
• Excellence Award in Creative Challenge Type SW R&D Program (2015)
• The National Scholarship for Science and Engineering (2011-2015)

---