arXiv subjects

Nikita Kazeev

Publications and source records attributed to Nikita Kazeev.

At least 19 recordsLinked to original sources

arXiv2026

Position: Align AI to Our Aspirations, Not Our Flaws

We argue that aligning AI to aggregated human preferences is the wrong target. With current technology, one can train AIs to share the values of a Silicon Valley techno-optimist, a degrowth environmentalist, a national-conservative culture warrior, a single-party state cadre, or a devout religious traditionalist. We should not. Human values produce societies that thrive or fail on the merits of those values - from failed states and extreme inequality to declining happiness, political polarization, and government dysfunction in the world's wealthiest democracies. The pluralistic-alignment program correctly diagnoses that there is no single "humanity" to align with, but is dangerous if taken as the main directive. We argue that AI should be trained to a non-negotiable floor of objective alignment goals - competence, bounded by the constraints of factual accuracy, honesty, and lawfulness and that pluralism belongs at the surface (language, register, conventions, missing-context defaults) and across the wide band of legitimate value tradeoffs that respect the floor, but not at the level of values that violate it. We highlight the empirical reality of unfiltered pluralistic values, propose four commitments as a constructive alternative, and engage six credible objections: commercial pressure and practical feasibility, democratic legitimacy, regulatory compliance, over-reliance on institutionalist explanations, the charge that the floor itself is culturally laden, and the limits of Coherent Extrapolated Volition.

cs.CY↗

arXiv2026

MiAD: Mirage Atom Diffusion for De Novo Crystal Generation

In recent years, diffusion-based models have demonstrated exceptional performance in searching for simultaneously stable, unique, and novel (S.U.N.) crystalline materials. However, most of these models don't have the ability to change the number of atoms in the crystal during the generation process, which limits the variability of model sampling trajectories. In this paper, we demonstrate the severity of this restriction and introduce a simple yet powerful technique, mirage infusion, which enables diffusion models to change the state of the atoms that make up the crystal from existent to non-existent (mirage) and vice versa. We show that this technique improves model quality by up to x2.5 compared to the same model without this modification. The resulting model, Mirage Atom Diffusion (MiAD), is an equivariant joint diffusion model for de novo crystal generation that is capable of altering the number of atoms during the generation process. MiAD achieves an 8.2% S.U.N. rate on the MP-20 dataset, which substantially exceeds existing state-of-the-art approaches. Code: https://github.com/andrey-okhotin/miad.git

cs.LG↗

arXiv2026

Crys-JEPA: Accelerating Crystal Discovery via Embedding Screening and Generative Refinement

De novo crystal generation seeks to discover materials that are not merely realistic, but also stable and novel. However, most existing generative models are trained to maximize the likelihood of observed crystals, which encourages samples to stay close to known materials yet not necessarily align with the criteria that matter in discovery. Our empirical analysis shows that current crystal generative models exhibit a clear conflict between stability and novelty: samples near the observed distribution tend to retain stability but offer limited novelty, whereas samples farther from it often lose stability rapidly. This suggests that the useful region for discovering crystals that are both stable and novel is extremely narrow. To move beyond this limitation, we introduce Crys-JEPA, a joint embedding predictive architecture for crystals that learns an energy-aware latent space preserving formation-energy differences. In this space, stability assessment can be reformulated as an embedding-based comparison against accessible training crystals, reducing the reliance on expensive energy evaluation and task-specific external references. Building on Crys-JEPA, we further develop a screening-and-refinement pipeline that identifies promising generated crystals and reintroduces them to refine the generative model. On MP-20 and Alex-MP-20 datasets, we achieve improvements over baselines up to 53.8% and 72.7% on V.S.U.N. metric, respectively.

cs.LG↗

arXiv2026

Composable Crystals: Controllable Materials Discovery via Concept Learning

De novo crystal generation, a central task in materials discovery, aims to generate crystals that are simultaneously valid, stable, unique, and novel. Existing methods mainly rely on black-box stochastic sampling, providing limited control over how generated structures move beyond the observed distribution. In this paper, we introduce a concept-based compositional framework for crystal generation. We train a vector-quantized variational autoencoder to automatically discover a shared set of reusable crystal concepts, which serve as building blocks for guided generation. These learned concepts naturally exhibit interpretability from both local atomic environments and global symmetry patterns, and generalize to crystals from different distributions. By recombining such concepts, our framework enables controllable exploration of novel crystals beyond the training distribution, rather than relying solely on unconstrained random sampling. To further improve composition efficiency, we introduce a composition generator and iteratively refine it using high-quality samples generated by the model itself. The resulting concept compositions are then used to condition downstream crystal generation. Numerical experiments on MP-20 and Alex-MP-20 show that compositing concepts separately increase base model up to 53.2% and 51.7% on V.S.U.N metric, with particular gains in novelty.

cs.LG↗

arXiv2026

Generative design of inorganic materials

Materials discovery is fundamental to advance next-generation technologies as well as for sustainable and circular economy. Beyond computational screening, generative models are efficient at finding materials with desired properties, via multi-modal learning using multiscale data. This perspective examines the landscape of generative design for inorganic materials and discusses the integration of multi-modal learning with high-throughput experimental validation. We contextualize these challenges through the lens of a generative design framework as a unified approach to address the data-driven inverse design of functional materials. The central idea of the framework is constructed around a foundation AI model for inorganic materials interlinked deeply with various property databases and high-throughput experiments via a machine learning driven closed loop, which enables the framework to solve key challenges in functional materials. We argue that domain-specific implementations of such integrated workflows represent a promising pathway toward the unresolved challenge of data-driven inverse design for atom-engineered inorganic functional materials.

cond-mat.mtrl-sci↗

arXiv2026

LeMat-GenBench: A Unified Evaluation Framework for Crystal Generative Models

Generative machine learning (ML) models hold great promise for accelerating materials discovery through the inverse design of inorganic crystals, enabling an unprecedented exploration of chemical space. Yet, the lack of standardized evaluation frameworks makes it challenging to evaluate, compare, and further develop these ML models meaningfully. In this work, we introduce LeMat-GenBench, a unified benchmark for generative models of crystalline materials, supported by a set of evaluation metrics designed to better inform model development and downstream applications. We release both an open-source evaluation suite and a public leaderboard on Hugging Face, and benchmark 12 recent generative models. Results reveal that an increase in stability leads to a decrease in novelty and diversity on average, with no model excelling across all dimensions. Altogether, LeMat-GenBench establishes a reproducible and extensible foundation for fair model comparison and aims to guide the development of more reliable, discovery-oriented generative models for crystalline materials.

cs.LG↗

arXiv2025

AI4X Roadmap: Artificial Intelligence for the advancement of scientific pursuit and its future directions

Artificial intelligence and machine learning are reshaping how we approach scientific discovery, not by replacing established methods but by extending what researchers can probe, predict, and design. In this roadmap we provide a forward-looking view of AI-enabled science across biology, chemistry, climate science, mathematics, materials science, physics, self-driving laboratories and unconventional computing. Several shared themes emerge: the need for diverse and trustworthy data, transferable electronic-structure and interatomic models, AI systems integrated into end-to-end scientific workflows that connect simulations to experiments and generative systems grounded in synthesisability rather than purely idealised phases. Across domains, we highlight how large foundation models, active learning and self-driving laboratories can close loops between prediction and validation while maintaining reproducibility and physical interpretability. Taken together, these perspectives outline where AI-enabled science stands today, identify bottlenecks in data, methods and infrastructure, and chart concrete directions for building AI systems that are not only more powerful but also more transparent and capable of accelerating discovery in complex real-world environments.

physics.soc-ph↗

arXiv2025

Energy Underprediction from Symmetry in Machine-Learning Interatomic Potentials

Machine learning interatomic potentials (MLIAPs) have emerged as powerful tools for accelerating materials simulations with near-density functional theory (DFT) accuracy. However, despite significant advances, we identify a critical yet overlooked issue undermining their reliability: a systematic energy underprediction. This problem becomes starkly evident in large-scale thermodynamic stability assessments. By performing over 12 million calculations using nine MLIAPs for over 150,000 inorganic crystals in the Materials Project, we demonstrate that most frontier models consistently underpredict energy above hull (Ehull), a key metric for thermodynamic stability, total energy, and formation energy, despite the fact that over 90\% of test structures (DFT-relaxed) are in the training data. The mean absolute errors (MAE) for Ehull exceed ~30 meV/atom even by the best model, directly challenging claims of achieving ``DFT accuracy'' for property predictions central to materials discovery, especially related to (meta-)stability. Crucially, we trace this underprediction to insufficient handling of symmetry degrees of freedom (DOF), constituting both lattice symmetry and Wyckoff site symmetries for the space group. MLIAPs exhibit pronounced errors (MAE for Ehull $>$ ~40 meV/atom) in structures with high symmetry DOF, where subtle atomic displacements significantly impact energy landscapes. Further analysis also indicates that the MLIAPs show severe energy underprediction for a large proportion of near-hull materials. We argue for improvements on symmetry-aware models such as explicit DOF encoding or symmetry-regularized loss functions, and more robust MLIAPs for predicting crystal properties where the preservation and breaking of symmetry are pivotal.

cond-mat.mtrl-sci↗

arXiv2025

Wyckoff Transformer: Generation of Symmetric Crystals

Crystal symmetry plays a fundamental role in determining its physical, chemical, and electronic properties such as electrical and thermal conductivity, optical and polarization behavior, and mechanical strength. Almost all known crystalline materials have internal symmetry. However, this is often inadequately addressed by existing generative models, making the consistent generation of stable and symmetrically valid crystal structures a significant challenge. We introduce WyFormer, a generative model that directly tackles this by formally conditioning on space group symmetry. It achieves this by using Wyckoff positions as the basis for an elegant, compressed, and discrete structure representation. To model the distribution, we develop a permutation-invariant autoregressive model based on the Transformer encoder and an absence of positional encoding. Extensive experimentation demonstrates WyFormer's compelling combination of attributes: it achieves best-in-class symmetry-conditioned generation, incorporates a physics-motivated inductive bias, produces structures with competitive stability, predicts material properties with competitive accuracy even without atomic coordinates, and exhibits unparalleled inference speed.

cond-mat.mtrl-sci↗

arXiv2023

The LHCb ultra-fast simulation option, Lamarr: design and validation

Detailed detector simulation is the major consumer of CPU resources at LHCb, having used more than 90% of the total computing budget during Run 2 of the Large Hadron Collider at CERN. As data is collected by the upgraded LHCb detector during Run 3 of the LHC, larger requests for simulated data samples are necessary, and will far exceed the pledged resources of the experiment, even with existing fast simulation options. An evolution of technologies and techniques to produce simulated samples is mandatory to meet the upcoming needs of analysis to interpret signal versus background and measure efficiencies. In this context, we propose Lamarr, a Gaudi-based framework designed to offer the fastest solution for the simulation of the LHCb detector. Lamarr consists of a pipeline of modules parameterizing both the detector response and the reconstruction algorithms of the LHCb experiment. Most of the parameterizations are made of Deep Generative Models and Gradient Boosted Decision Trees trained on simulated samples or alternatively, where possible, on real data. Embedding Lamarr in the general LHCb Gauss Simulation framework allows combining its execution with any of the available generators in a seamless way. Lamarr has been validated by comparing key reconstructed quantities with Detailed Simulation. Good agreement of the simulated distributions is obtained with two-order-of-magnitude speed-up of the simulation phase.

hep-ex↗

arXiv2022

Robust Neural Particle Identification Models

The volume of data processed by the Large Hadron Collider experiments demands sophisticated selection rules typically based on machine learning algorithms. One of the shortcomings of these approaches is their profound sensitivity to the biases in training samples. In the case of particle identification (PID), this might lead to degradation of the efficiency for some decays not present in the training dataset due to differences in input kinematic distributions. In this talk, we propose a method based on the Common Specific Decomposition that takes into account individual decays and possible misshapes in the training data by disentangling common and decay specific components of the input feature set. We show that the proposed approach reduces the rate of efficiency degradation for the PID algorithms for the decays reconstructed in the LHCb detector.

hep-ex↗

arXiv2022

Unveiling the complex structure-property correlation of defects in 2D materials based on high throughput datasets

Modification of physical properties of materials and design of materials with on-demand characteristics is at the heart of modern technology. Rare application relies on pure materials--most devices and technologies require careful design of materials properties through alloying, creating heterostructures of composites or controllable introduction of defects. At the same time, such designer materials are notoriously difficult for modelling. Thus, it is very tempting to apply machine learning methods for such systems. Unfortunately, there is only a handful of machine learning-friendly material databases available these days. We develop a platform for easy implementation of machine learning techniques to materials design and populate it with datasets on pristine and defected materials. Here we describe datasets of defects in represented 2D materials such as MoS2, WSe2, hBN, GaSe, InSe, and black phosphorous, calculated using DFT. Our study provides a data-driven physical understanding of complex behaviors of defect properties in 2D materials, holding promise for a guide to the development of efficient machine learning models. In addition, with the increasing enrollment of datasets, our database could provide a platform for designing of materials with predetermined properties.

cond-mat.mtrl-sci↗

arXiv2022

Generative models uncertainty estimation

In recent years fully-parametric fast simulation methods based on generative models have been proposed for a variety of high-energy physics detectors. By their nature, the quality of data-driven models degrades in the regions of the phase space where the data are sparse. Since machine-learning models are hard to analyse from the physical principles, the commonly used testing procedures are performed in a data-driven way and can't be reliably used in such regions. In our work we propose three methods to estimate the uncertainty of generative models inside and outside of the training phase space region, along with data-driven calibration techniques. A test of the proposed methods on the LHCb RICH fast simulation is also presented.

cs.LG↗

arXiv2022

Towards Reliable Neural Generative Modeling of Detectors

The increasing luminosities of future data taking at Large Hadron Collider and next generation collider experiments require an unprecedented amount of simulated events to be produced. Such large scale productions demand a significant amount of valuable computing resources. This brings a demand to use new approaches to event generation and simulation of detector responses. In this paper, we discuss the application of generative adversarial networks (GANs) to the simulation of the LHCb experiment events. We emphasize main pitfalls in the application of GANs and study the systematic effects in detail. The presented results are based on the Geant4 simulation of the LHCb Cherenkov detector.

physics.ins-det↗

arXiv2020

Muon identification for LHCb Run 3

Muon identification is of paramount importance for the physics programme of LHCb. In the upgrade phase, starting from Run 3 of the LHC, the trigger of the experiment will be solely based on software. The luminosity increase to $2\times10^{33}$ cm$^{-2}$s$^{-1}$ will require an improvement of the muon identification criteria, aiming at performances equal or better than those of Run 2, but in a much more challenging environment. In this paper, two new muon identification algorithms developed in view of the LHCb upgrade are presented, and their performance in terms of signal efficiency versus background reduction is shown.

hep-ex↗

arXiv2019

Machine Learning on data with sPlot background subtraction

Data analysis in high energy physics often deals with data samples consisting of a mixture of signal and background events. The sPlot technique is a common method to subtract the contribution of the background by assigning weights to events. Part of the weights are by design negative. Negative weights lead to the divergence of some machine learning algorithms training due to absence of the lower bound in the loss function. In this paper we propose a mathematically rigorous way to train machine learning algorithms on data samples with background described by sPlot to obtain signal probabilities conditioned on observables, without encountering negative event weight at all. This allows usage of any out-of-the-box machine learning methods on such data.

cs.LG↗

arXiv2019

Machine Learning on sWeighted Data

Data analysis in high energy physics has to deal with data samples produced from different sources. One of the most widely used ways to unfold their contributions is the sPlot technique. It uses the results of a maximum likelihood fit to assign weights to events. Some weights produced by sPlot are by design negative. Negative weights make it difficult to apply machine learning methods. The loss function becomes unbounded. This leads to divergent neural network training. In this paper we propose a mathematically rigorous way to transform the weights obtained by sPlot into class probabilities conditioned on observables, thus enabling to apply any machine learning algorithm out-of-the-box.

hep-ex↗

arXiv2019

Fast Data-Driven Simulation of Cherenkov Detectors Using Generative Adversarial Networks

The increasing luminosities of future Large Hadron Collider runs and next generation of collider experiments will require an unprecedented amount of simulated events to be produced. Such large scale productions are extremely demanding in terms of computing resources. Thus new approaches to event generation and simulation of detector responses are needed. In LHCb, the accurate simulation of Cherenkov detectors takes a sizeable fraction of CPU time. An alternative approach is described here, when one generates high-level reconstructed observables using a generative neural network to bypass low level details. This network is trained to reproduce the particle species likelihood function values based on the track kinematic parameters and detector occupancy. The fast simulation is trained using real data samples collected by LHCb during run 2. We demonstrate that this approach provides high-fidelity results.

physics.ins-det↗