📢📢 News
-
Our paper “Latent Debate” has been accepted at NeurIPS 2026
-
Our survey about the role of small models has been accepted by Computational Linguistics
-
A small demo about how AI is changing teaching
-
Our benchmark AgentComa has been accepted at ACL 2026
-
Our query-level uncertainty has been accepted at ICLR 2026
My name is Lihu Chen (陈立虎), and I am currently an ISFP Researcher (permanent position) at Inria Saclay in the CEDAR team.
Previously, I was a Research Associate at Imperial College London.
Before that, I did a one-year postdoc at Inria Saclay. I obtained my PhD at DIG team of Télécom Paris, which is a member of Institut Polytechnique de Paris. I was co-supervised by Fabian Suchanek and Gaël Varoquaux.
My research primarily focuses on natural language processing (NLP) and large language models (LLMs). I am dedicated to developing efficient, reliable, and open-source models and tools, with a particular emphasis on information extraction and biomedical applications. Specifically, my research topics include:
- Large vs Small: Collaborative AI Systems (Frugal AI). In our survey, we systematically examine the collaborations between LLMs and SMs. My research will build upon this foundational concept, focusing on leveraging the complementary strengths of both LLMs and SMs.
- Generation-Free Signals in Foundation Models. I am interested in using internal signals of models to make them more efficient and trustworthy. The question here is how we can find useful signals before generating any answer tokens.
- Efficient Knowledge-Augmented LLMs. LLMs may struggle with tasks that require domain-specific expertise or up-to-date information. I aim to explore cost-effective methods for enhancing reasoning capabilities, e.g. logical reasoning and medical applications.
- Long-term Research Question: Can AI Develop Human-like Reasoning Capabilities—or Go Beyond Them?: How can we develop AI systems that go beyond statistical patterns? New Architectures? Multimodality? New Learning Objectives? Neuro-symbolic Integration? We’ll see!
Misc
- My Github
- Human Language: Chinese Mandarin (Native); English (Fluent); French (Basic)
- Travel Photos
Contact
Email:[firstname].[lastname][AT]imperial.ac.uk
I am open to discussions and collaborations. Feel free to reach out!