Jinyi Liu (刘金毅)

💻

Jinyi Liu

Ph.D. Candidate

About Me

Jinyi Liu (刘金毅) is currently pursuing a Ph.D. at Tianjin University under the supervision of Professor Jianye Hao, as a member of the DRL Lab. His research interests primarily focus on DRL, LLMs, and LLM Agents. His work aims to explore synergies between decision-making frameworks and language-based AI systems, advancing applications in autonomous reasoning and human-AI collaboration.

🔬 TJU DRL-LAB (by Jianye Hao, Yan Zheng and Hongyao Tang) is seeking collaborators (interns, MS/PhD)! 👋 DM me if you're interested!

Interests

Deep Reinforcement Learning
LLM Post-training (Scaling and RL)
LLM and LLM Agents
AI for Science

Education

PhD (and MSc)
Tianjin University
BSc
NorthEastern University

News

📝 2026-01 Paper accepted by ACM TheWebConf 2026 Industry: AFE-Master.
🔥 2025-12 Released our beginner-friendly LLM Agent tutorial (website & PDF).
🎤 2025-12 The 137th RLCHINA Paper Seminar hosted!
📝 2025-11 1 paper accepted by AAAI 2025 (MADC)!
📝 2025-09 1 paper accepted by NeurIPS 2025.
📝 2025-07 1 paper accepted by SCALR@COLM (Atomic Reasoner).
📝 2025-07 1 paper accepted by ICCV 2025 (RoboAnnotatorX)!
📝 2025-06 1 paper accepted by ICML 2025 Workshop MAS (MADC)!
📝 2025-05 3 papers accepted by ACL 2025 (long paper, Atomic Reasoner, DualRAG, WoT)!
🏆 2025-02 Nominated as Distinguished PC Member of AAMAS 2025.

Featured Publications

LLM Post-Training (TTS)

From Chaos to Order: The Atomic Reasoner Framework for Fine-grained Reasoning in Large Language Models

A lightweight large language model inference framework that performs structured and fine-grained natural language reasoning without the need for complex search and external tools.

Aug 1, 2025

LLM Agent

Cellagent: An llm-driven multi-agent framework for automated single-cell data analysis

An LLM-driven multi-agent framework for single-cell data analysis, ensuring high-quality results with minimal effort.

Aug 1, 2024

DRL

Ovd-explorer: Optimism should not be the sole pursuit of exploration in noisy environments

We propose Optimistic Value Distribution Explorer (OVD-Explorer) to achieve a noise-aware optimistic exploration for continuous control.

May 1, 2024

Recent Publications

Hebin Liang, Jianye Hao, Jinyi Liu, Yi Ma, Zilin Cao, Jing Liang, Kun Shao, Zhaocheng Du, Fei Ni, Yifu Yuan, Yan Zheng (2026). AFE-Master: Enhancing LLM-Driven Autonomous Feature Engineering with Domain-Specific Language Parsing and Guided Local Search. ACM TheWebConf 2026 Industry.

Cite

Shuyue Hu, Siyue Ren, Yang Chen, Chunjiang Mu, Jinyi Liu, Zhiyao Cui, Yiqun Zhang, Hao Li, Dongzhan Zhou, Jia Xu, Others (2025). Hands-on LLM-based Agents: A Tutorial for General Audiences. TechRxiv.

PDF Cite Project Source Document