We introduce ChemPro, a progressive benchmark with 4100 natural language question-answer pairs in Chemistry, across 4 coherent sections of difficulty designed to assess the proficiency of Large Language Models (LLMs) in a broad spectrum of general chemistry topics. We include Multiple Choice Questions and Numerical Questions spread across fine-grained information recall, long-horizon reasoning, multi-concept questions, problem-solving with nuanced articulation, and straightforward questions in a balanced ratio, effectively covering Bio-Chemistry, Inorganic-Chemistry, Organic-Chemistry and Physical-Chemistry. ChemPro is carefully designed analogous to a student's academic evaluation for basic to high-school chemistry. A gradual increase in the question difficulty rigorously tests the ability of LLMs to progress from solving basic problems to solving more sophisticated challenges. We evaluate 45+7 state-of-the-art LLMs, spanning both open-source and proprietary variants, and our analysis reveals that while LLMs perform well on basic chemistry questions, their accuracy declines with different types and levels of complexity. These findings highlight the critical limitations of LLMs in
To enhance large language models (LLMs) for chemistry problem solving, several LLM-based agents augmented with tools have been proposed, such as ChemCrow and Coscientist. However, their evaluations are narrow in scope, leaving a large gap in understanding the benefits of tools across diverse chemistry tasks. To bridge this gap, we develop ChemToolAgent, an enhanced chemistry agent over ChemCrow, and conduct a comprehensive evaluation of its performance on both specialized chemistry tasks and general chemistry questions. Surprisingly, ChemToolAgent does not consistently outperform its base LLMs without tools. Our error analysis with a chemistry expert suggests that: For specialized chemistry tasks, such as synthesis prediction, we should augment agents with specialized tools; however, for general chemistry questions like those in exams, agents' ability to reason correctly with chemistry knowledge matters more, and tool augmentation does not always help.
The spatial distribution of the chemical reservoirs in protoplanetary disks is key to elucidate the composition of planets, especially habitable ones. However, the partitioning of the main elements among the refractory and volatile phases is still elusive. Key parameters such as the carbon-to-oxygen C/O elemental ratio and the ionization fraction remain poorly constrained, with the latter potentially orders of magnitude lower than in the interstellar medium. Moreover, the thermal structure of the gas is also poorly known, despite its deep influence on gas-phase chemistry. In this context, ortho-to-para ratios could provide selective and sensitive probes. Recent ALMA observations have measured the spatially resolved column density of ortho-and para-H2CO in the transition disk orbiting TW Hya and derived the radial profile of the ortho-to-para ratio. Yet, current disk models do not include the nuclear-spin-resolved chemistry required to interpret these observations. The present work aims to fill this gap, by combining a parametric disk physical model of TW Hya with the UGAN network, updated to include a comprehensive description of the nuclear-spin-resolved chemistry of formaldehyde.
Multimodal scientific reasoning remains a significant challenge for large language models (LLMs), particularly in chemistry, where problem-solving relies on symbolic diagrams, molecular structures, and structured visual data. Here, we systematically evaluate 40 proprietary and open-source multimodal LLMs, including GPT-5, o3, Gemini-2.5-Pro, and Qwen2.5-VL, on a curated benchmark of Olympiad-style chemistry questions drawn from over two decades of U.S. National Chemistry Olympiad (USNCO) exams. These questions require integrated visual and textual reasoning across diverse modalities. We find that many models struggle with modality fusion, where in some cases, removing the image even improves accuracy, indicating misalignment in vision-language integration. Chain-of-Thought prompting consistently enhances both accuracy and visual grounding, as demonstrated through ablation studies and occlusion-based interpretability. Our results reveal critical limitations in the scientific reasoning abilities of current MLLMs, providing actionable strategies for developing more robust and interpretable multimodal systems in chemistry. This work provides a timely benchmark for measuring progress in
We develop a structural theory of chirality for inverse semigroups and show how it propagates canonically to étale groupoids and twisted groupoid $C^*$-algebras. Starting from inverse semigroup data equipped with admissible twist information, we construct a canonical twisted universal groupoid in the sense of Paterson and introduce a mirror correspondence encoding intrinsic asymmetry. Our main result identifies a structural obstruction to mirror self-duality at the level of twisted universal groupoids and shows that this obstruction descends to an obstruction for the associated reduced twisted groupoid $C^*$-algebra to be isomorphic to its opposite. The framework is representation-independent, yet compatible with concrete germ groupoid models, and provides a unified bridge between partial symmetries, groupoid structures, and analytic invariants in noncommutative operator algebras.
High expansion velocity carbon stars (HVCs) are a rare class of evolved stars whose circumstellar envelopes (CSEs) combine C-rich chemistry with unusually high expansion velocities typical of O-rich massive evolved stars. AFGL2233 has been proposed as a high-mass evolved object that exhausted hot-bottom burning. Studying its chemistry is essential to understand the nature and evolution of these objects. We characterize the chemical composition and isotopic ratios of the CSE of AFGL2233 and investigate chemical peculiarities, including the presence of N- and O-bearing species in a C-rich environment. We carried out a complete line survey at 3 mm and 1 mm using the IRAM 30m telescope, complemented by Herschel/HIFI FIR observations and interferometric maps of SiO, C2H, and HCN. Molecular emission was analyzed using rotational diagrams and radiative transfer modeling under the LVG approximation. Column densities and fractional abundances were derived for more than 30 molecular species, including isotopologues, and compared with other evolved stars. The Gaia DR3 distance of 1.236 kpc implies a luminosity of ~2 Lsun, consistent with an initial mass of 4.5-9 Msun. The molecular inventory
Efficient chemical kinetic model inference and application in combustion are challenging due to large ODE systems and widely separated time scales. Machine learning techniques have been proposed to streamline these models, though strong nonlinearity and numerical stiffness combined with noisy data sources make their application challenging. Here, we introduce ChemKANs, a novel neural network framework with applications both in model inference and simulation acceleration for combustion chemistry. ChemKAN's novel structure augments the generic Kolmogorov Arnold Network Ordinary Differential Equations (KAN-ODEs) with knowledge of the information flow through the relevant kinetic and thermodynamic laws. This chemistry-specific structure combined with the expressivity and rapid neural scaling of the underlying KAN-ODE algorithm instills in ChemKANs a strong inductive bias, streamlined training, and higher accuracy predictions compared to standard benchmarks, while facilitating parameter sparsity through shared information across all inputs and outputs. In a model inference investigation, we benchmark the robustness of ChemKANs to sparse data containing up to 15% added noise, and superfl
The optimization-based damage detection and damage state digital twinning capabilities are examined here of a novel conditional-labeled generative adversarial network methodology. The framework outperforms current approaches for fault anomaly detection as no prior information is required for the health state of the system: a topic of high significance for real-world applications. Specifically, current artificial intelligence-based digital twinning approaches suffer from the uncertainty related to obtaining poor predictions when a low number of measurements is available, physics knowledge is missing, or when the damage state is unknown. To this end, an unsupervised framework is examined and validated rigorously on the benchmark structural health monitoring measurements of Z24 Bridge: a post-tensioned concrete highway bridge in Switzerland. In implementing the approach, firstly, different same damage-level measurements are used as inputs, while the model is forced to converge conditionally to two different damage states. Secondly, the process is repeated for a different group of measurements. Finally, the convergence scores are compared to identify which one belongs to a different da
This paper presents a measurement of the production cross-section of a $Z$ boson in association with $b$- or $c$-jets, in proton-proton collisions at $\sqrt{s} = 13$ TeV with the ATLAS experiment at the Large Hadron Collider using data corresponding to an integrated luminosity of 140 fb$^{-1}$. Inclusive and differential cross-sections are measured for events containing a $Z$ boson decaying into electrons or muons and produced in association with at least one $b$-jet, at least one $c$-jet, or at least two $b$-jets with transverse momentum $p_\textrm{T} > 20$ GeV and rapidity $|y| < 2.5$. Predictions from several Monte Carlo generators based on next-to-leading-order matrix elements interfaced with a parton-shower simulation, with different choices of flavour schemes for initial-state partons, are compared with the measured cross-sections. The results are also compared with novel predictions, based on infrared and collinear safe jet flavour dressing algorithms. Selected $Z + \ge 1 c$-jet observables, optimized for sensitivity to intrinsic-charm, are compared with benchmark models with different intrinsic-charm fractions.
Three-body recombination, or ternary association, is a termolecular reaction in which three particles collide, forming a bound state between two, whereas the third escapes freely. Three-body recombination reactions play a significant role in many systems relevant to physics and chemistry. In particular, they are relevant in cold and ultracold chemistry, quantum gases, astrochemistry, atmospheric physics, physical chemistry, and plasma physics. As a result, three-body recombination has been the subject of extensive work during the last 50 years, although primarily from an experimental perspective. Indeed, a general theory for three-body recombination remains elusive despite the available experimental information. Our group recently developed a direct approach based on classical trajectory calculations in hyperspherical coordinates for three-body recombination to amend this situation, leading to a first principle explanation of ion-atom-atom and atom-atom-atom three-body recombination processes. This review aims to summarize our findings on three-body recombination reactions and identify the remaining challenges in the field.
Gait abnormality detection is critical for the early discovery and progressive tracking of musculoskeletal and neurological disorders, such as Parkinson's and Cerebral Palsy. Especially, analyzing the foot-floor contacts during walking provides important insights into gait patterns, such as contact area, contact force, and contact time, enabling gait abnormality detection through these measurements. Existing studies use various sensing devices to capture such information, including cameras, wearables, and force plates. However, the former two lack force-related information, making it difficult to identify the causes of gait health issues, while the latter has limited coverage of the walking path. In this study, we leverage footstep-induced structural vibrations to infer foot-floor contact profiles and detect gait abnormalities. The main challenge lies in modeling the complex force transfer mechanism between the foot and the floor surfaces, leading to difficulty in reconstructing the force and contact profile during foot-floor interaction using structural vibrations. To overcome the challenge, we first characterize the floor vibration for each contact type (e.g., heel, midfoot, and
The Structural theory of chemistry introduces chemical/molecular structure as a combination of relative arrangement and bonding patterns of atoms in molecule. Nowadays, the structure of atoms in molecules is derived from the topological analysis of the quantum theory of atoms in molecules (QTAIM). In this context a molecular structure is varied by large geometrical variations and concomitant reorganization of electronic structure that are usually taking place in chemical reactions or under extreme hydrostatic pressure. In this report a new mode of structural variation is introduced within the context of the newly proposed multi-component QTAIM (MC-QTAIM) that originates from mass variation of nuclei. Accordingly, XCN and CNX series of species are introduced where X stands for a quantum particle with a unit of positive charge and a variable mass that is varied in discrete steps between the masses of proton and positron. Ab initio non-Born-Oppenheimer (non-BO) calculations are done on both series of species and the resulting non-BO wavefunctions are used for the MC-QTAIM analysis revealing a triatomic structure for the proton mass and a diatomic structure for the positron mass. In bo
We present two open-source implementations of the Locally Optimal Block Preconditioned Conjugate Gradient (LOBPCG) algorithm to find a few eigenvalues and eigenvectors of large, possibly sparse matrices. We then test LOBPCG for various quantum chemistry problems, encompassing medium to large, dense to sparse, wellbehaved to ill-conditioned ones, where the standard method typically used is Davidson's diagonalization. Numerical tests show that, while Davidson's method remains the best choice for most applications in quantum chemistry, LOBPCG represents a competitive alternative, especially when memory is an issue, and can even outperform Davidson for ill-conditioned, non diagonally dominant problems.
A nuclear radius of $^{22}$C is investigated with the total reaction cross sections at medium- to high-incident energies in order to resolve the radius puzzle in which two recent interaction cross section measurements using $^1$H and $^{12}$C targets show the quite different radii. The cross sections of $^{22}$C are calculated consistently for these target nuclei within a reliable microscopic framework, the Glauber theory. To describe appropriately such a reaction involving a spatially extended nucleus, the multiple scattering processes within the Glauber theory are fully taken into account, that is, the multi-dimensional integration in the Glauber amplitude is evaluated using a Monte Carlo technique without recourse to the optical-limit approximation. We discuss the sensitivity of the spatially extended halo tail to the total reaction cross sections. The root-mean-square matter radius obtained in this study is consistent with that extracted from the recent cross section measurement on $^{12}$C target. We show that the simultaneous reproduction of the two recent measured cross sections is not feasible within this framework.
Accurate structural response prediction forms a main driver for structural health monitoring and control applications. This often requires the proposed model to adequately capture the underlying dynamics of complex structural systems. In this work, we utilize a learnable Extended Kalman Filter (EKF), named the Neural Extended Kalman Filter (Neural EKF) throughout this paper, for learning the latent evolution dynamics of complex physical systems. The Neural EKF is a generalized version of the conventional EKF, where the modeling of process dynamics and sensory observations can be parameterized by neural networks, therefore learned by end-to-end training. The method is implemented under the variational inference framework with the EKF conducting inference from sensing measurements. Typically, conventional variational inference models are parameterized by neural networks independent of the latent dynamics models. This characteristic makes the inference and reconstruction accuracy weakly based on the dynamics models and renders the associated training inadequate. In this work, we show that the structure imposed by the Neural EKF is beneficial to the learning process. We demonstrate the
Astrochemistry lies at the nexus of astronomy, chemistry, and molecular physics. On the basis of precise laboratory data, a rich collection of more than 200 familiar and exotic molecules have been identified in the interstellar medium, the vast majority by their unique rotational fingerprint. Despite this large body of work, there is scant evidence in the radio band for the basic building blocks of chemistry on earth -- five and six-membered rings -- despite long standing and sustained efforts during the past 50 years. In contrast, a peculiar structural motif, highly unsaturated carbon in a chain-like arrangement, is instead quite common in space. The recent astronomical detection of cyanobenzene, the simplest aromatic nitrile, in the dark molecular cloud TMC-1, and soon afterwards in additional pre-stellar, and possibly protostellar sources, establishes that aromatic chemistry is likely widespread in the earliest stages of star formation. The subsequent discovery of cyanocyclopentadienes and even cyanonapthlenes in TMC-1 provides further evidence that organic molecules of considerable complexity are readily synthesized in regions with high visual extinction but where the low tempe
This Report summarises the results of the second year's activities of the LHC Higgs Cross Section Working Group. The main goal of the working group was to present the state of the art of Higgs Physics at the LHC, integrating all new results that have appeared in the last few years. The first working group report Handbook of LHC Higgs Cross Sections: 1. Inclusive Observables (CERN-2011-002) focuses on predictions (central values and errors) for total Higgs production cross sections and Higgs branching ratios in the Standard Model and its minimal supersymmetric extension, covering also related issues such as Monte Carlo generators, parton distribution functions, and pseudo-observables. This second Report represents the next natural step towards realistic predictions upon providing results on cross sections with benchmark cuts, differential distributions, details of specific decay channels, and further recent developments.
This work presents a combined experimental and theoretical investigation of the nuclear reaction $^{6}$Li + $^{12}$C at a laboratory energy of 68 MeV. The reaction products are identified via the standard $Δ$E-E technique. Angular distributions are constructed for the elastic, inelastic, and deuteron transfer channels by detecting emitted particles -- $^{6}$Li and $α$. Elastic and inelastic scattering of $^{6}$Li on $^{12}$C are analyzed using both the Optical Model and Coupled channels approaches, with the interaction described by a double-folding potential. This potential is calculated based on the three-body wave function of $^{6}$Li. Pronounced coupled-channel effects are observed, which modify the potential and allow accurate reproduction of the experimental cross sections. The resulting polarized potentials provide a more precise description of the initial-state interaction for further reaction modelling. The deuteron transfer channel, $^{12}$C($^{6}$Li, $α$)$^{14}$N, is studied using the Coupled Reaction Channels method. The coupling between the transfer and elastic channels is implemented using the three-body wave function of $^{6}$Li. As an alternative, a regular wave func
We study regular irreducible inclusions $B\subset A$ of simple unital $C^*$-algebras admitting a conditional expectation. We introduce a generalized notion of quasi-basis extending Watatani's framework and show that such inclusions admit a unitary orthonormal generalized quasi-basis. As a consequence, we prove that every regular irreducible inclusion in this setting is canonically isomorphic to a reduced twisted crossed product of $B$ by its Weyl group. This extends earlier crossed product characterizations beyond the finite-index setting.
The stereochemistry of 6s2 (E) lone pair of divalent Pb and trivalent Bi (PbII and BiIII designated by M*) in structurally related PbO, PbFX (X= Cl, Br, I), BiOX (X= F, Cl, Br, I) and Bi2NbO5F is rationalized. The lone pair LP presence determined by its sphere of influence E, equal to those of oxygen or fluorine anions, was settled by its center then giving M*-E directions and distances. Detailed description of structural features of both elements in the title compounds characterized by [PbEO]n and [BiEO]n layers allowed to show the evolution of M*-E distance versus the changes with the square pyramidal SP coordination polyhedra. All are different, in red PbO one finds {PbEO4E4} square antiprism, a {[Bi.E]O4X4Xapical} monocapped square antiprism in PbFX and BiOX and {BiEO4F4}square antiprism in Bi2NbO5F. To analyze the crystal chemistry results, the electronic structures of these compounds were calculated within density functional theory DFT. Real space analyses of electron localization illustrate a full volume development of the lone pair on PbII within {PbEO4E4} in PbOE, {PbEF4X4} in PbFXE and Bi(III) within {BiEO4X4} square antiprisms, contrary to Bi(III) within {[Bi.E]O4F4Fapic