Research

UE for NN models and transformers. Also the second branch of my PhD thesis and my main research interest area at the moment:

• UNCERTAINTY-LINE: Length-invariant estimation of uncertainty for large language models, EMNLP 2025
• Fact-checking the output of large language models via token-level uncertainty quantification, ACL 2024
• Uncertainty estimation of transformer predictions for misclassification detection, ACL 2022
• How certain is your Transformer?, EACL 2021
• Dropout strikes back: Improved uncertainty estimation via diversity sampling, AIST 2021
• Active learning and uncertainty estimation (chapter), Machine Learning Meets Quantum Physics, 2020
• Deeper connections between neural networks and Gaussian processes speed-up active learning, IJCAI 2019
• Dropout-based active learning for regression, AIST/COPA 2018

Exciting discoveries on the fusion of machine learning, theoretical physics, semiconductors, and density functional theory. Also the main branch of my PhD thesis. Main works:

• Phonon stability boundary and deep elastic strain engineering of lattice thermal conductivity, PNAS 2024
• Machine learning for deep elastic strain engineering of semiconductor electronic band structure and effective mass, npj Computational Materials 2021
• Elastic strain engineering of materials, US Patent
• Metallization of diamond, PNAS 2020
• Deep elastic strain engineering of bandgap through machine learning, PNAS 2019
  
  

Finite-difference research I started my scientific life from. Works:

• Compact difference scheme for parabolic and Schrödinger-type equations with variable coefficients, JCP 2018
• A fourth-order accurate difference scheme for a differential equation with variable coefficients, Mathematical Models and Computer Simulations 2018
• Compact high-order difference approximations for rod lateral vibrations equation, 2017
• Compact difference scheme for the differential equation with piecewise-constant coefficient, Mat.Mod. 2017
• Compact difference schemes for weakly-nonlinear parabolic and Schrodinger-type equations and systems, 2017
• 4th order difference scheme for the differential equation with variable coefficients, Mat.Mod. 2017
• Compact difference schemes for the diffusion and Schrödinger equations. Approximation, stability, convergence, effectiveness, monotony, JCM 2014

Mostly ML papers related to work and not related to the topics above.

• ClimateGPT: Towards AI Synthesizing Interdisciplinary Research on Climate Change, 2024
• Building a Classification System for Failed Test Reports: Industrial Experience, AITest 2020
• Building an Adaptive Logs Classification System: Industrial Report, TMPA 2019
• User-assisted log analysis for quality control of distributed fintech applications, AITest 2019
• Churn prediction for game industry based on cohort classification ensemble, EEML 2017

• International Scientific Committee Member, IOAI 2024, 2025, 2026
• PC Member, ICDM 2025
• Organizing Committee Member, ICDM 2024
• PC Member, LREC-COLING 2024
• Proceedings Chair, ML Track Chair, PC Member, AIST 2023, 2024
• ML Track Chair, PC Member, AIST 2021
• PC Member, IJCAI 2021
• PC Member, AIST 2020
• PC Member, EEML 2019
• Organizing Committee Member, AIST 2018
• Curator, AISTHack 2018
• Reviewer, Transactions on Semiconductor Manufacturing
• Reviewer, Applied Numerical Mathematics
• Organizing Committee Member, AIST 2016
• Organizing Committee Member, RuSSIR 2016
• Curator, SNA Hackaton 2016
• Curator, GoToHack2 2016