I am a Computer Science Ph.D. candidate at University of Illinois Urbana-Champaign advised by Prof. Jimeng Sun. I am also working closely with Prof. Jiawei Han, Prof. Sheng Wang, Dr. Lijun Wu, Dr. Kun Qian and Tian Wang. Before that, I earned my B.Eng and MS degree in Automation at Tsinghua University.

My research focuses on foundation models (e.g., large language models (LLMs), multi-modal models), reinforcement learning (RL), and agents, with their applications in healthcare, biomedicine, and recommendation systems. I am particularly interested in how these models can be used for reasoning, retrieval, and decision-making in complex, real-world scenarios.

• LLM-centric Research: Search Agents (DeepRetrieval, s3), Knowledge Graphs (GenRES), LLM Efficiency (Cascade Speculative Drafting), Revisiting SFT Forgetting (TALR)
• LLMs/RL + Healthcare: EHR Reasoning Model with RL (EHRMind), Clinical Trial Foundation Model (Panacea, TrialPanorama)
• BioMedical Foundation Models: Molecule Foundation Model (Pisces, R²-DDI), Single-cell Foundation Model (EpiAgent)
• LLMs/RL + Recommendation: Bridge RecSys with LLMs using RL (Rec-R1, Shop-R1)

I am currently seeking research-oriented internship opportunities for the summer of 2026. Please contact me if you have any recruitment openings.

News

01/2026 Two papers SFT Doesn’t Always Hurt General Capabilities: Revisiting Domain-Specific Fine-Tuning in LLMs and Shop-R1 are accepted by ICLR 2026.

12/2025 Our Survey Adaptation of Agentic AI is released! We categorized existing agent papers into four paradigms, which makes the overall research landscape much clearer.

10/2025 Our Rec-R1: Bridging Generative Large Language Models and User-Centric Recommendation Systems via Reinforcement Learning is accepted by TMLR.

09/2025 Our EpiAgent: Foundation model for single-cell epigenomic data is accepted by Nature Methods (IF=32.1).

08/2025 Starting Machine Learning Research Internship at Keiji AI.

08/2025 Our s3: You Don’t Need That Much Data to Train a Search Agent via RL is accepted by EMNLP 2025.

07/2025 Our DeepRetrieval: Hacking Real Search Engines and Retrievers with Large Language Models via Reinforcement Learning is accepted by COLM 2025.

07/2025 Our GT2Vec: Large Language Models for Knowledge Graph Augmented Text Embedding is accepted by KDD 2025 Workshop SKnow-LLM.

05/2025 Starting Applied Scientist Internship at Amazon SFAI, Palo Alto, advised by Hyokun Yun and Lihong Li.

04/2025 Our Pisces: A multi-modal data augmentation approach for drug combination synergy prediction is accepted by Cell Genomics.

09/2024 Our Cascade Speculative Drafting for Even Faster LLM Inference is accepted by NeurIPS 2024.

05/2024 Starting Applied Scientist Internship at Amazon, Palo Alto, advised by Sheng Wang and Danai Koutra.

03/2024 Our GENRES: Rethinking Evaluation for Generative Relation Extraction in the Era of Large Language Models is accepted by NAACL 2024.

01/2024 Our CAMBranch: Contrastive Learning with Augmented MILPs for Branching is accepted by ICLR 2024.

Selected Papers

• Jiacheng Lin, Zhongruo Wang, Kun Qian, Tian Wang, Arvind Srinivasan, Hansi Zeng, Ruocheng Jiao, Xie Zhou, Jiri Gesi, Dakuo Wang, Yufan Guo, Kai Zhong, Weiqi Zhang, Sujay Sanghavi, Changyou Chen, Hyokun Yun, Lihong Li, SFT Doesn’t Always Hurt General Capabilities: Revisiting Domain-Specific Fine-Tuning in LLMs, ICLR, 2026. [PDF]
• Jiacheng Lin, Tian Wang, Kun Qian, Rec-R1: Bridging Generative Large Language Models and User-Centric Recommendation Systems via Reinforcement Learning, TMLR, 2025. [PDF][CODE]
• Pengcheng Jiang*, Jiacheng Lin∗, Lang Cao, Runchu Tian, SeongKu Kang, Zifeng Wang, Jimeng Sun, Jiawei Han, DeepRetrieval: Hacking Real Search Engines and Retrievers with Large Language Models via Reinforcement Learning, COLM, 2025. [PDF][CODE]
• Hanwen Xu*, Jiacheng Lin∗, Addie Woicik, Zixuan Liu, Jianzhu Ma, Sheng Zhang, Hoifung Poon, Liewei Wang, Sheng Wang, Pisces: A multi-modal data augmentation approach for drug combination synergy prediction, Cell Genomics, 2025. [PDF][CODE]
• Jiacheng Lin∗, Hanwen Xu∗, Addie Woicik, Jianzhu Ma and Sheng Wang, Pisces: A cross-modal contrastive learning approach to synergistic drug combination prediction, RECOMB, 2023. [PDF][CODE]
• Jiacheng Lin, Lijun Wu, Jinhua Zhu, Xiaobo Liang, Yingce Xia, Shufang Xie, Tao Qin and Tie-Yan Liu, R²-DDI: Relation-aware Feature Refinement for Drug-Drug Interaction Prediction, Briefings in Bioinformatics, Volume 24, Issue 1, 2023. [PDF][CODE]