arXiv subjects

Nazar Buzun

Publications and source records attributed to Nazar Buzun.

15 recordsLinked to original sources

arXiv2026

Plan, Don't Pose: Long Composite Motion Generation with Text-Aligned BFM

Text-to-motion (T2M) generation has broad applications in character animation, virtual avatars, and human-robot interaction. Existing methods typically generate pose trajectories or motion tokens directly from language, forcing a single model to handle semantic interpretation, long-horizon structure, and low-level physical realization. This coupling makes them costly and often unreliable for long, compositional, or semantically dense prompts. We propose Text2BFM, the first framework that aligns natural language with pretrained Behavioral Foundation Models (BFMs) for T2M generation without relying on heavy end-to-end motion generators. Text2BFM operates in the latent policy space of a frozen BFM, using it as an executable motion prior. A text-aligned variational behavioral bottleneck compresses BFM policy-latent sequences into compact motion representations that are compatible with language and preserve long-horizon behavioral structure. Generation is performed in this compact behavioral manifold with a lightweight conditional generator, and the resulting latent encoded behaviors are decoded into policy latents that drive the pretrained frozen BFM. By decoupling semantic planning from motion execution, Text2BFM achieves efficient, robust T2M generation and strong performance on long, compositional textual descriptions.

cs.LG↗

arXiv2026

Q-RAG: Long Context Multi-step Retrieval via Value-based Embedder Training

Retrieval-Augmented Generation (RAG) methods enhance LLM performance by efficiently filtering relevant context for LLMs, reducing hallucinations and inference cost. However, most existing RAG methods focus on single-step retrieval, which is often insufficient for answering complex questions that require multi-step search. Recently, multi-step retrieval approaches have emerged, typically involving the fine-tuning of small LLMs to perform multi-step retrieval. This type of fine-tuning is highly resource-intensive and does not enable the use of larger LLMs. In this work, we propose Q-RAG, a novel approach that fine-tunes the Embedder model for multi-step retrieval using reinforcement learning (RL). Q-RAG offers a competitive, resource-efficient alternative to existing multi-step retrieval methods for open-domain question answering and achieves state-of-the-art results on the popular long-context benchmarks BabiLong and RULER for contexts up to 10M tokens. Code is available at https://github.com/griver/Q-RAG

cs.LG↗

arXiv2026

Unlocking the Duality between Flow and Field Matching

Conditional Flow Matching (CFM) unifies conventional generative paradigms such as diffusion models and flow matching. Interaction Field Matching (IFM) is a newer framework that generalizes Electrostatic Field Matching (EFM) rooted in Poisson Flow Generative Models (PFGM). While both frameworks define generative dynamics, they start from different objects: CFM specifies a conditional probability path in data space, whereas IFM specifies a physics-inspired interaction field in an augmented data space. This raises a basic question: are CFM and IFM genuinely different, or are they two descriptions of the same underlying dynamics? We show that they coincide for a natural subclass of IFM that we call forward-only IFM. Specifically, we construct a bijection between CFM and forward-only IFM. We further show that general IFM is strictly more expressive: it includes EFM and other interaction fields that cannot be realized within the standard CFM formulation. Finally, we highlight how this duality can benefit both frameworks: it provides a probabilistic interpretation of forward-only IFM and yields novel, IFM-driven techniques for CFM.

cs.LG↗

arXiv2025

HOTA: Hamiltonian framework for Optimal Transport Advection

Optimal transport (OT) has become a natural framework for guiding the probability flows. Yet, the majority of recent generative models assume trivial geometry (e.g., Euclidean) and rely on strong density-estimation assumptions, yielding trajectories that do not respect the true principles of optimality in the underlying manifold. We present Hamiltonian Optimal Transport Advection (HOTA), a Hamilton-Jacobi-Bellman based method that tackles the dual dynamical OT problem explicitly through Kantorovich potentials, enabling efficient and scalable trajectory optimization. Our approach effectively evades the need for explicit density modeling, performing even when the cost functionals are non-smooth. Empirically, HOTA outperforms all baselines in standard benchmarks, as well as in custom datasets with non-differentiable costs, both in terms of feasibility and optimality.

cs.LG↗

arXiv2024

ENOT: Expectile Regularization for Fast and Accurate Training of Neural Optimal Transport

We present a new approach for Neural Optimal Transport (NOT) training procedure, capable of accurately and efficiently estimating optimal transportation plan via specific regularization on dual Kantorovich potentials. The main bottleneck of existing NOT solvers is associated with the procedure of finding a near-exact approximation of the conjugate operator (i.e., the c-transform), which is done either by optimizing over non-convex max-min objectives or by the computationally intensive fine-tuning of the initial approximated prediction. We resolve both issues by proposing a new, theoretically justified loss in the form of expectile regularisation which enforces binding conditions on the learning process of dual potentials. Such a regularization provides the upper bound estimation over the distribution of possible conjugate potentials and makes the learning stable, completely eliminating the need for additional extensive fine-tuning. Proposed method, called Expectile-Regularised Neural Optimal Transport (ENOT), outperforms previous state-of-the-art approaches on the established Wasserstein-2 benchmark tasks by a large margin (up to a 3-fold improvement in quality and up to a 10-fold improvement in runtime). Moreover, we showcase performance of ENOT for varying cost functions on different tasks such as image generation, showing robustness of proposed algorithm. OTT-JAX library includes our implementation of ENOT algorithm https://ott-jax.readthedocs.io/en/latest/tutorials/ENOT.html

cs.LG↗

arXiv2024

Align Your Intents: Offline Imitation Learning via Optimal Transport

Offline Reinforcement Learning (RL) addresses the problem of sequential decision-making by learning optimal policy through pre-collected data, without interacting with the environment. As yet, it has remained somewhat impractical, because one rarely knows the reward explicitly and it is hard to distill it retrospectively. Here, we show that an imitating agent can still learn the desired behavior merely from observing the expert, despite the absence of explicit rewards or action labels. In our method, AILOT (Aligned Imitation Learning via Optimal Transport), we involve special representation of states in a form of intents that incorporate pairwise spatial distances within the data. Given such representations, we define intrinsic reward function via optimal transport distance between the expert's and the agent's trajectories. We report that AILOT outperforms state-of-the art offline imitation learning algorithms on D4RL benchmarks and improves the performance of other offline RL algorithms by dense reward relabelling in the sparse-reward tasks.

cs.LG↗

arXiv2022

Explain My Surprise: Learning Efficient Long-Term Memory by Predicting Uncertain Outcomes

In many sequential tasks, a model needs to remember relevant events from the distant past to make correct predictions. Unfortunately, a straightforward application of gradient based training requires intermediate computations to be stored for every element of a sequence. This requires to store prohibitively large intermediate data if a sequence consists of thousands or even millions elements, and as a result, makes learning of very long-term dependencies infeasible. However, the majority of sequence elements can usually be predicted by taking into account only temporally local information. On the other hand, predictions affected by long-term dependencies are sparse and characterized by high uncertainty given only local information. We propose MemUP, a new training method that allows to learn long-term dependencies without backpropagating gradients through the whole sequence at a time. This method can potentially be applied to any recurrent architecture. LSTM network trained with MemUP performs better or comparable to baselines while requiring to store less intermediate data.

cs.LG↗

arXiv2021

Landmarks Augmentation with Manifold-Barycentric Oversampling

The training of Generative Adversarial Networks (GANs) requires a large amount of data, stimulating the development of new augmentation methods to alleviate the challenge. Oftentimes, these methods either fail to produce enough new data or expand the dataset beyond the original manifold. In this paper, we propose a new augmentation method that guarantees to keep the new data within the original data manifold thanks to the optimal transport theory. The proposed algorithm finds cliques in the nearest-neighbors graph and, at each sampling iteration, randomly draws one clique to compute the Wasserstein barycenter with random uniform weights. These barycenters then become the new natural-looking elements that one could add to the dataset. We apply this approach to the problem of landmarks detection and augment the available annotation in both unpaired and in semi-supervised scenarios. Additionally, the idea is validated on cardiac data for the task of medical segmentation. Our approach reduces the overfitting and improves the quality metrics beyond the original data outcome and beyond the result obtained with popular modern augmentation methods.

cs.CV↗

arXiv2021

Gaussian approximation for penalized Wasserstein barycenters

In this work we consider regularized Wasserstein barycenters (average in Wasserstein distance) in Fourier basis. We prove that random Fourier parameters of the barycenter converge to some Gaussian random vector by distribution. The convergence rate has been derived in finite-sample case with explicit dependence on measures count ($n$) and the dimension of parameters ($p$).

math.ST↗

arXiv2021

Strong Gaussian Approximation for the Sum of Random Vectors

This paper derives a new strong Gaussian approximation bound for the sum of independent random vectors. The approach relies on the optimal transport theory and yields \textit{explicit} dependence on the dimension size $p$ and the sample size $n$. This dependence establishes a new fundamental limit for all practical applications of statistical learning theory. Particularly, based on this bound, we prove approximation in distribution for the maximum norm in a high-dimensional setting ($p >n$).

math.ST↗

arXiv2021

BRULÈ: Barycenter-Regularized Unsupervised Landmark Extraction

Unsupervised retrieval of image features is vital for many computer vision tasks where the annotation is missing or scarce. In this work, we propose a new unsupervised approach to detect the landmarks in images, validating it on the popular task of human face key-points extraction. The method is based on the idea of auto-encoding the wanted landmarks in the latent space while discarding the non-essential information (and effectively preserving the interpretability). The interpretable latent space representation (the bottleneck containing nothing but the wanted key-points) is achieved by a new two-step regularization approach. The first regularization step evaluates transport distance from a given set of landmarks to some average value (the barycenter by Wasserstein distance). The second regularization step controls deviations from the barycenter by applying random geometric deformations synchronously to the initial image and to the encoded landmarks. We demonstrate the effectiveness of the approach both in unsupervised and semi-supervised training scenarios using 300-W, CelebA, and MAFL datasets. The proposed regularization paradigm is shown to prevent overfitting, and the detection quality is shown to improve beyond the state-of-the-art face models.

cs.CV↗

arXiv2020

Change-point detection in high-dimensional covariance structure

In this paper we introduce a novel approach for an important problem of break detection. Specifically, we are interested in detection of an abrupt change in the covariance structure of a high-dimensional random process -- a problem, which has applications in many areas e.g., neuroimaging and finance. The developed approach is essentially a testing procedure involving a choice of a critical level. To that end a non-standard bootstrap scheme is proposed and theoretically justified under mild assumptions. Theoretical study features a result providing guaranties for break detection. All the theoretical results are established in a high-dimensional setting (dimensionality $p \gg n$). Multiscale nature of the approach allows for a trade-off between sensitivity of break detection and localization. The approach can be naturally employed in an on-line setting. Simulation study demonstrates that the approach matches the nominal level of false alarm probability and exhibits high power, outperforming a recent approach.

math.ST↗

arXiv2020

Unsupervised non-parametric change point detection in quasi-periodic signals

We propose a new unsupervised and non-parametric method to detect change points in intricate quasi-periodic signals. The detection relies on optimal transport theory combined with topological analysis and the bootstrap procedure. The algorithm is designed to detect changes in virtually any harmonic or a partially harmonic signal and is verified on three different sources of physiological data streams. We successfully find abnormal or irregular cardiac cycles in the waveforms for the six of the most frequent types of clinical arrhythmias using a single algorithm. The validation and the efficiency of the method are shown both on synthetic and on real time series. Our unsupervised approach reaches the level of performance of the supervised state-of-the-art techniques. We provide conceptual justification for the efficiency of the method and prove the convergence of the bootstrap procedure theoretically.

cs.LG↗

arXiv2017

Bootstrap for change point detection

In Change point detection task Likelihood Ratio Test (LRT) is sequentially applied in a sliding window procedure. Its high values indicate changes of parametric distribution in the data sequence. Correspondingly LRT values require predefined bound for their maximum. The maximum value has unknown distribution and may be calibrated with multiplier bootstrap. Bootstrap procedure convolves independent components of the Likelihood function with random weights, that enables to estimate empirically LRT distribution. For this empirical distribution of the LRT we show convergence rates to the real maximum value distribution.

math.ST↗

arXiv2017

Stochastic Analysis in Problems, part 1 (in Russian)

This book contains a large number of exercises related to different stochastic disciplines. Difficulty of the problems varies from the basic level in the first chapter up to the analysis of articles in Probability, Statistics and Computer Science. Range of the given exercises covers most of the significant classical results of the area including techniques that arise in modern stochastic disciplines. The book is recommended for the course "Stochastic Analysis in problems" which is held by the Department of Mathematical Foundations of Control in Moscow Institute of Physics and Technology. Content of the book may be used for practical classes, exams and for self-education reasons in various stochastic disciplines.

math.PR↗