opened — Apr 2026

Linlin Jia

— Ph.D. in Machine Learning

Linlin Jia, Ph.D., Machine Learning Research Scientist
that's me

Research Scientist · Postdoc University of Bern, Switzerland

work done

today's research

Graph Machine
Learning
Graph Representation
Learning
Spatio-Temporal
Learning
Graph AI for
Science & Industry
LLM Systems
& Agents

can graphs + LLM agents do real science or industry work?

open to opportunities

Open to Work

fast onboarding

Roles
ML Research Scientist ML Engineer Applied Scientist AI Scientist Research Scientist
Locations
CH · EU (on-site / hybrid) China · NA (open to relocation)
Employment
Full-time (preferred) Consulting
Contact me for a role Download CV

About Me

Building graph representation learning across domains: open-source tools, models, and AI systems for research & industry.

Now Postdoc · PRG, University of Bern · graph representation learning
Domains drug discovery (redox), Swiss hydrology, engineering diagrams (PLANALYSER)
Industry N-Banker (AI strategy / LLM agents) · WATTELSE (HVAC audit)
OSS graphkit-learn · liulian · redoxprediction
Mentor 20+ students supervised
Funded SNSF · Innosuisse · ANR · UniBE-funded projects
Past PhD '21 LITIS Lab, INSA Rouen · MSc/BSc XJTU
Looking ML Research Scientist / Engineer roles · academia & industry
9
Publications
8+
Projects
5+
Grants
15+
Collaborators
20+
Students Mentored
8+
Industry Experience
130
Citations
6
H-index

Selected Publications

130+ citations · Google Scholar

Spatio-temporal transformer architecture for Swiss river water temperature forecasting
ICPR
Conference • 2026 · Accepted
Benchmarking Transformers on Spatio-Temporal River Water Temperature Modeling
L. Jia, B. Fankhauser, V. Bigler, K. Riesen
International Conference on Pattern Recognition (ICPR) 2026: accepted
In press
Code
Graph-based redox potential prediction pipeline (Journal of Computational Chemistry 2024)
J. Comp. Chem.
Journal • 2024
Predicting Redox Potentials by Graph-Based Machine Learning Methods
L. Jia, É. Brémond, L. Zaida, B. Gaüzère, V. Tognetti, L. Joubert
Journal of Computational Chemistry, 2024
9 citations
DOI Preprint Code
Graph embedding-and-classification learning framework bridging distinct graph spaces (ACPR 2023)
ACPR
Conference • 2023
Bridging Distinct Spaces in Graph-based Machine Learning
L. Jia, X. Ning, B. Gaüzère, P. Honeine, K. Riesen
Asian Conference on Pattern Recognition (ACPR) 2023, Kitakyushu, Japan
5 citations
DOI PDF Code
Epi-DNNs: epidemiological-prior-informed deep neural network for COVID-19 dynamics (Computers in Biology and Medicine 2023)
CompBioMed
Journal • 2023
Epi-DNNs: Epidemiological Priors Informed Deep Neural Networks for Modeling COVID-19 Dynamics
X. Ning, L. Jia, Y. Wei, X.-A. Li, F. Chen
Computers in Biology and Medicine, 2023
61 citations
DOI
Stability analysis of graph edit distance heuristics: experimental results (Electronics 2022)
Electronics
Journal • 2022
A Study on the Stability of Graph Edit Distance Heuristics
L. Jia, V. Tognetti, L. Joubert, B. Gaüzère, P. Honeine
Electronics, 2022
2 citations
DOI Preprint
Graph representations used by graph kernels based on linear patterns (Expert Systems with Applications 2022)
ESWA
Journal • 2022
Graph Kernels Based on Linear Patterns: Theoretical and Experimental Comparisons
L. Jia, B. Gaüzère, P. Honeine
Expert Systems with Applications, 2022
25 citations
DOI Preprint Code
graphkit-learn accuracy benchmark across graph datasets (Pattern Recognition Letters 2021)
PRL
Journal • 2021
graphkit-learn: A Python Library for Graph Kernels Based on Linear Patterns
L. Jia, B. Gaüzère, P. Honeine
Pattern Recognition Letters, 2021
14 citations
DOI Preprint Code
Graph pre-image method based on graph edit distances: conceptual overview (SSPR 2021)
SSPR
Workshop • 2021
A Graph Pre-image Method Based on Graph Edit Distances
L. Jia, B. Gaüzère, P. Honeine
IAPR Joint International Workshops SPR & SSPR 2020: Springer LNCS, 2021
9 citations
DOI Preprint Slides Video
Metric learning framework for graph edit costs applied to regression (SSPR 2021)
SSPR
Workshop • 2021
A Metric Learning Approach to Graph Edit Costs for Regression
L. Jia, B. Gaüzère, F. Yger, P. Honeine
IAPR Joint International Workshops SPR & SSPR 2020: Springer LNCS, 2021
7 citations
DOI Preprint Slides Video
Google Patents page for China patent CN106376041B on ELM-based SDN mobility prediction
Patent
Patent • 2016
A Kind of Name Data Network Mobility Switching Method Predicted Using ELM
Qu Hua, Zhao Jihong, Wu Jinkang, Jia Linlin, et al.
China Patent CN106376041B
Patent
Ph.D. Dissertation · 2021

Bridging graph and kernel spaces: a pre-image perspective

L. Jia · LITIS Lab, INSA Rouen Normandie, France

Graph kernels, graph edit distances, and pre-image algorithms and applications. Supervised by Prof. Paul Honeine and Prof. Benoit Gaüzère.

Thesis PDF Defense slides Defense video

Projects

Open the full App Gallery

Research Outline

Graph Machine Learning

Bridging graph kernels, edit distances, GNNs, transformers. Pre-image and generation problems.

Spatio-Temporal Learning

Time series with entity embeddings, spatial graphs for environmental forecasting.

Graph AI for science and industry

Molecular property prediction (foundations for drug discovery), computational chemistry, smart engineering, hydrology, digital humanities, finance, healthcare.

LLM Systems & Agents

RAG pipelines, fine-tuning (LoRA), agent orchestration for production AI.

Graph Neural Networks Deep Learning LLMs PyTorch Computational Chemistry Transformers Graph Kernels Time Series Chemoinformatics Edit Distance Pre-image Forecasting Spatio-temporal scikit-learn Fine-tuning Molecules Entity Embeddings Computer Vision Document Analysis Agents Networks Hydrology Smart Engineering Digital Humanities Graphs Drug Discovery Foundation Models Diffusion Models Polymers Redox Potentials Deployment MLOps Inference Docker

Professional Experience

2025 - 2026

Postdoctoral Researcher

Pattern Recognition Group (PRG), University of Bern, Switzerland

SNSF-funded research on spatio-temporal graph convolutional networks and LLMs for Swiss river temperature forecasting. Developing transformer architectures with entity embeddings and graph structures. Paper accepted by ICPR 2026.

2024 - 2025

Scientific Collaborator

iCoSys Institute, HES-SO Fribourg, Switzerland

Applied topology-aware graph and vision deep learning for symbol detection, connection prediction, and graph construction in engineering diagrams and handwritten document analysis. Contributed to SNSF-funded Virtual Bodmer project (accepted) for 3D digital reconstruction of ancient papyrus.

2023 - 2024

Research Fellow

Pattern Recognition Group (PRG), University of Bern, Switzerland

SNSF-funded research on novel graph matching algorithms. Published at ACPR 2023 on bridging distinct spaces in graph-based machine learning.

2021 - 2022

Postdoctoral Researcher

COBRA Lab, INSA Rouen Normandie, France

Applied graph ML to polymer optimization (OCTOPUSSY project) and redox potential prediction. Published in Journal of Computational Chemistry (2024).

2017 - 2021

Ph.D. in Computer Science

LITIS Lab, INSA Rouen Normandie, France

Thesis: "Bridging graph and kernel spaces: a pre-image perspective": Graph kernels, graph edit distances, and pre-image problems.

Thesis PDF Defense slides Defense video
2014 - 2017

M.Sc. in Software Engineering

Xi'an Jiaotong University, China

Service-oriented programmable control for Software-Defined Networks; applied extreme learning machine (ELM) to network mobility prediction. Resulted in 1 China patent (CN106376041B).

2010 - 2014

B.S. in Information Engineering

Xi'an Jiaotong University, China

Foundations in signal processing, networks, and software engineering.

Skills & Technologies

ML/AI
GNNs LLMs Transformers Graph Kernels Graph Edit Distances Pre-image Agents Vision Transformer Computer Vision Deep Learning Machine Learning Time series scikit-learn YOLO
Programming
Python PyTorch C++ Java JavaScript Cython MATLAB Spring Boot React.js
Tools & Infra
Docker FastAPI Git Cloud Linux CI/CD HPC LaTeX
Domain
RDKit DeepChem Gaussian Chemoinformatics Hydrology Spatio-Temporal Analysis
AI Tools
GitHub Copilot Claude Dify vLLM ollama
Languages
Chinese (Native) English (Fluent) French (B2)

Selected Awards & Grants

Swiss National Science Foundation (SNSF/SNF)

SNSF Grants × 3

Swiss National Science Foundation · 2020 – 2026

Graph matching (188496), handwriting recognition (217594), and spatio-temporal river forecasting (206352).

University of Bern

UniBE ECSC Open Round 2023/2

Faculty of Science, University of Bern · 2024 – 2026

Early Career Scientist Cooperation. Applicant & Principal Investigator. Predicting molecular energies with multiscale graph ML built on Interacting Quantum Atoms (IQA) and GNNs.

Innosuisse: Swiss Innovation Agency Innosuisse: Swiss Innovation Agency

Innosuisse Grant

Swiss Innovation Agency · 2024 – 2025

PLANALYSER: AI-driven HVAC-concept audit (with WATTELSE AG).

Agence Nationale de la Recherche (ANR) Agence Nationale de la Recherche (ANR)

ANR Grant (PhD project APi)

Agence Nationale de la Recherche · 2018 – 2021

PhD thesis funding (LITIS, INSA Rouen), project Apprivoiser la Pré-image: graph kernels & ML for structured data.

Academic Services

Supervision

  • 20+ students mentored
  • Topics: Computer vision, Graph-based learning, Smart engineering, Deep learning, LLMs, Agent systems

Associations

  • Member of the Swiss Association for Pattern Recognition (SAPR) (2024-)
  • Member of the Marie Curie Alumni Association China Chapter (2024)
  • Associated member of the LITIS lab (2022)

Reviewing

  • International Conference on Pattern Recognition 2024

Invited Talks

  • 🇫🇷 Invited speaker, GRAPHADON Summer School, Rouen, France — Bridging Graph Spaces: Novel Algorithms and Their Applications (Jun 2024).
  • 🇯🇵 Oral presentation, ACPR 2023, Kitakyushu, Japan (Nov 2023).
  • 🇫🇷 Video presentations at S+SSPR Workshops 2020 — A Graph Pre-image Method Based on Graph Edit Distances [YT · Bilibili] & A Metric Learning Approach to Graph Edit Costs for Regression [YT · Bilibili] (Jan 2021).
  • 🇩🇪🇫🇷 Invited talk, 2020 Online CSC-Seminar in DE. and FR. — A Graph Pre-image Method Based on Graph Edit Distances (awarded; Sep 2020) [Slides · Certificate].

News & Updates

Apr 2026
Neobanker is live at #InnoEX! One of the first agentic digital financial information platforms.
Apr 2026
Paper accepted at ICPR 2026! Spatio-temporal transformers for Swiss river temperature prediction.
Jan 2026
Started the cooperation with N-Banker, leading AI strategy and LLM agent development.
2025
SNSF Virtual Bodmer project accepted - AI for 3D papyrus reconstruction.
Jun 2024
Invited speaker at GRAPHADON Summer School, Rouen, France.
2024
Paper published in J. Computational Chemistry - Graph-based redox potential prediction.
Nov 2023
Oral presentation at ACPR 2023, Kitakyushu, Japan.
2022
Paper published in Expert Systems with Applications - Kernel pattern comparisons.
2021
Ph.D. thesis defended at INSA Rouen, France. Defense video

Contact

Email

linlin.jia@unibe.ch

jajupmochi@gmail.com

Location

🇨🇭 University of Bern · Pattern Recognition Group

Schützenmattstrasse 14, 3012 Bern, Switzerland

Phone

🇨🇭 +41 78 224 1419 (Switzerland)

🇨🇳 +86 <please-add> (China)

Open in Maps