Research projects of the laboratory (the page is under construction).
Fair DivisionFair division problems arise when indivisible resources—such as property, computational resources, or tasks—must be allocated among participants with different preferences. […]
Efficient serving and performance modeling of LLMs in industrial scenariosReal-world application scenarios of large language models, including those within cloud services, differ substantially from laboratory settings due to the […]
Authorship attribution algorithmsEvery author has unconscious linguistic habits: how often they use function words, which combinations of letters and word forms they […]
Developing methods, data, and tools of modern applied computational linguistics for less-resourced languages.Preparing and evaluating language data; building lexical-semantic, morphological, and syntactic tools based on both classical and neural approaches; transferring annotations […]
Path, Trajectory, and Motion Planning for Modern RoboticsRobotics is one of the most rapidly advancing areas of science and engineering. In particular, current research is increasingly focused […]
Hard benchmarks for general-purpose large language modelsFinding, theoretically analyzing, and constructing tasks that remain hard even for “strong” language models and that expose their limitations, which […]
Masked diffusion language modelsThe study of the fundamental properties of masked diffusion language models — a new class of models in which text […]
Boolean Satisfiability Problems: Algorithms and Computational ComplexityThe Boolean satisfiability problem (SAT) asks whether there exists an assignment of truth values to the variables that satisfies a […]
Analysis of neural activity in Danio rerioRecent advances in neuroscience have made it possible to record the activity of nearly all neurons in certain model organisms, […]
Multi-Agent PathfindingThe Multi-Agent Pathfinding (MAPF) problem consists of finding a set of conflict-free paths on a graph. On the one hand, […]
Efficient Sequence ModelingResearch on architectures for efficient long-sequence processing, primarily state-space models and their hybrids with transformers. We investigate how the design […]
Applying LLMs to tasks that often arise in a software engineer’s work: analyzing and annotating code, automatically constructing custom architectural representations of the developer’s codebase, and moreThis includes effective agentic harnesses: how to make “modest tier” coding LLMs solve programming tasks at the level of larger […]
Machine Learning and Heuristic Search for Hard Combinatorial ProblemsImagine that you need to design an efficient algorithm that can solve a Rubik’s Cube from any scrambled state using […]