I am a research scientist at Yandex Research, based at the lab’s HSE University branch. I earned my Ph.D. at the Centre for Deep Learning and Bayesian Methods.

My early work was on probabilistic machine learning — approximate inference, generative models, and structured prediction. Over the past few years I have shifted toward 3D computer vision and neural rendering, where probabilistic intuition translates naturally into differentiable, uncertainty-aware approaches to scene understanding. I also teach a graduate 3D Computer Vision course at HSE University. In my spare time I dabble in computer graphics and play table tennis.

Twitter / CV / Google Scholar / Github

Selected Publications

• MADrive: Memory-Augmented Driving Scene Modeling. Polina Karpikova, Daniil Selikhanovych, Kirill Struminsky*, Ruslan Musaev, Maria Golitsyna, and Dmitry Baranchuk. CVPR 2026 (Findings).
• CasTex: Cascaded Text-to-Texture Synthesis via Explicit Texture Maps and Physically-Based Shading. Mishan Aliev, Dmitry Baranchuk, and Kirill Struminsky. WACV 2026.
• Differentiable Rendering with Reparameterized Volume Sampling. Nikita Morozov, Denis Rakitin, Oleg Desheulin, Dmitry Vetrov, and Kirill Struminsky. AISTATS 2024 (also Neural Fields workshop at ICLR 2023).
• Leveraging Recursive Gumbel-Max Trick for Approximate Inference in Combinatorial Spaces. Kirill Struminsky, Artyom Gadetsky, Denis Rakitin*, Danil Karpushkin, and Dmitry Vetrov. NeurIPS 2021.
• Low-Variance Black-Box Gradient Estimates for the Plackett-Luce Distribution. Artyom Gadetsky, Kirill Struminsky, Christopher Robinson, Novi Quadrianto, and Dmitry Vetrov. AAAI 2020 (oral).
• The Deep Weight Prior. Modeling a Prior Distribution for CNNs Using Generative Models. Andrey Atanov, Senya Ashukha, Kirill Struminsky, Dmitry Vetrov, and Max Welling. ICLR 2019.
• Quantifying Learning Guarantees for Convex but Inconsistent Surrogates. Kirill Struminsky, Simone Lacoste-Julien, and Anton Osokin. NeurIPS 2018.