Author Image

Hi, I am Hongjin

Hongjin Qian

Postdoctoral Researcher at BAAI & PKU

I am Hongjin Qian, a postdoctoral researcher at BAAI and Peking University, working with Prof. Tiejun Huang and Dr. Zhongyuan Wang. I earned my PhD in 2024 from the Gaoling School of Artificial Intelligence (GSAI) at Renmin University of China, under the supervision of Prof. Zhicheng Dou and Prof. Ji-Rong Wen . I hold a Master’s degree from the University of Sydney (2019) and a Bachelor’s degree from Nankai University (2017). My experience includes research internships at Huawei and WeChat Group, as well as contributions to AI startups.

Academic
Coding
Engineering
Cooking
Photography
Guitar

Research

Search Agent
2025-Present
Agentic search is a retrieval approach where an AI agent actively plans, decomposes, and executes multi-step information-seeking actions to answer complex user questions.
Publications
Retrieval-Augmentation Generation
2022-Present
Retrieval-Augmented Generation (RAG) is a method that first retrieves relevant information from an external knowledge source and then combines it with the model’s input to generate more accurate and informative responses.
Publications
Conversational Search
2021-Present
Conversational search is an interactive search paradigm where users and systems engage in a dialogue, allowing queries, clarifications, and refinements across multiple turns to iteratively reach more accurate and context-aware results.
Publications
Others
2020-Present
Dialogue System, QA System, Ranking, Retrieval, Theory, etc.
Publications

Experiences

1
Assistant Research Fellow
Peking University

Aug 2025 - Present, Beijing, China

Responsibilities:
  • Memory-Enhanced Agent
  • Agentic Search

Postdoctoral Researcher
Peking University

Oct 2024 - Present, Beijing, China

Responsibilities:
  • Memory-Enhanced LLMs
  • Efficiency KV cache techniques
2

3
Research Intern -> Postdoctoral Researcher
Beijing Academy of Artificial Intelligence.

Nov 2023 - Present, Beijing, China

Beijing Academy of Artificial Intelligence (BAAI) is a non-profit research institute dedicated to promoting collaboration among academia and industries, as well as fostering top talents and a focus on long-term research on the fundamentals of AI technology.

Responsibilities:
  • Agentic Search
  • Memory-Enhanced LLMs
  • Rretrieval-Augmented Generation

Research Intern
Wechat Group, Tencent.

Jun 2023 - Oct 2023, Beijing, China

Responsibilities:
  • LLM for IR
  • LLM for QA
4

5
Research Intern
Poisson Lab, Huawei.

Apr 2022 - May 2023, Beijing, China

Responsibilities:
  • Pretraining for IR
  • Model-oriented IR

PhD. Researcher
Gaoling School of Artificial Intelligence, Renmin University of China.

Sept 2020 - Jun 2024, Beijing, China

The Gaoling School of Artificial Intelligence (GSAI) at Renmin University of China (RUC) is a prestigious institution dedicated to shaping the future of AI. GSAI has consistently ranked first in Information Retrieval worldwide, according to CSRankings, from 2022 to 2024.

Responsibilities:
  • Personalized AI
  • Conversational AI
  • Information Retrieval
6

7
Intern NLP Engineer
Beijing Academy of Artificial Intelligence.

Jun 2020 - Mar 2021, Beijing, China

Responsibilities:
  • QA System dedicated in Governance Domain
  • Dense Vector Search
  • Fine-Grained Named Entity Recognition

NLP Engineer
Elensdata.

Jan 2019 - Sept 2020, Beijing, China

Elensdata is a start-up company which offers high-calibre data science/AI solutions that help real businesses, in media, finance, etc.

Responsibilities:
  • Core NLP Toolkit for Chinese, English and other languages
  • NLP Applications in Multiple Domains (Financial, Security, Media etc.)
  • Large-Scale Pretrained Language Model and Text Generation
8
Patents Statistics
2019-2024

20 Patents in Total
18 Granted Patents
6 First-Inventor Patents

Academic Service
2021-2025

Reviewer / PC Member:
Neurips, ICLR, ICML, ACL, EMNLP, MM
EACL, ACL ARR, SIGKDD, theWebConf, TOIS

Hierarchical Memory-Enhanced Knowledge Reasoning for Large Language Models
Owner Jan 2026 - Dec 2028

This project focuses on exploring techniques to expand the knowledge scale and memory scale at the input stage. The goal is to overcome the limitations of current LLMs in complex knowledge reasoning, knowledge memorization, and global knowledge understanding. This will be achieved by constructing a hierarchical memory mechanism that enables the scaling, memorization, and dynamic, coordinated retrieval of multi-source, heterogeneous knowledge.

MemoRAG
Owner Aug 2024 - Present

MemoRAG is a next-generation retrieval-augmented generation system with long-term memory, enabling superior context-aware information retrieval and enhanced performance on complex tasks where traditional RAG systems struggle.

Infomatica
Owner Aug 2025 - Present

Informatica is a comprehensive collection of systematic research projects focused on deep research systems. Our mission is to provide open-source, scalable frameworks, datasets, data synthesis methods, models, and demonstrations.

A Semantic Parsing Method Based on Rules and Learning
First Inventor CN112347793A

This patent proposes a semantic parsing method that combines rules and learning-based approaches.