Search

The recent revolutionary advance in generative AI enables the generation of realistic and coherent texts by large language models (LLMs). Despite many existing evaluation metrics on the quality of the generated texts, there is still a lack of rigorous assessment of how well LLMs perform in complex and demanding writing assessments. This study examines essays generated by ten leading LLMs for the analytical writing assessment of the Graduate Record Exam (GRE). We assessed these essays using both human raters and the e-rater automated scoring engine as used in the GRE scoring pipeline. Notably, the top-performing Gemini and GPT-4o received an average score of 4.78 and 4.67, respectively, falling between "generally thoughtful, well-developed analysis of the issue and conveys meaning clearly" and "presents a competent analysis of the issue and conveys meaning with acceptable clarity" according to the GRE scoring guideline. We also evaluated the detection accuracy of these essays, with detectors trained on essays generated by the same and different LLMs., Comment: 20 pages, 6 figures

We present a calculation by including the relativistic and off-shell contributions to the interaction potentials between two spin-1/2 fermions mediated by the exchange of light spin-0 particles, in both momentum and coordinate spaces. Our calculation is based on the four-point Green function rather than the scattering amplitude. Among the sixteen potential components, eight that vanish in the non-relativistic limit are shown to acquire nonzero relativistic and off-shell corrections. In addition to providing relativistic and off-shell corrections to the operator basis commonly used in the literature, we introduce an alternative operator basis that facilitates the derivation of interaction potentials in coordinate space. Furthermore, we calculate both the long-range and short-range components of the potentials, which can be useful for future experimental analyses at both macroscopic and atomic scales.

The application of machine learning methodologies for predicting properties within materials science has garnered significant attention. Among recent advancements, Kolmogorov-Arnold Networks (KANs) have emerged as a promising alternative to traditional Multi-Layer Perceptrons (MLPs). This study evaluates the impact of substituting MLPs with KANs within three established machine learning frameworks: Allegro, Neural Equivariant Interatomic Potentials (NequIP), and the Edge-Based Tensor Prediction Graph Neural Network (ETGNN). Our results demonstrate that the integration of KANs generally yields enhanced prediction accuracies. Specifically, replacing MLPs with KANs in the output blocks leads to notable improvements in accuracy and, in certain scenarios, also results in reduced training times. Furthermore, employing KANs exclusively in the output block facilitates faster inference and improved computational efficiency relative to utilizing KANs throughout the entire model. The selection of an optimal basis function for KANs is found to be contingent upon the particular problem at hand. Our results demonstrate the strong potential of KANs in enhancing machine learning potentials and material property predictions.

Non-adiabatic molecular dynamics (NAMD) simulations have become an indispensable tool for investigating excited-state dynamics in solids. In this work, we propose a general framework, N$^2$AMD which employs an E(3)-equivariant deep neural Hamiltonian to boost the accuracy and efficiency of NAMD simulations. The preservation of Euclidean symmetry of Hamiltonian enables N$^2$AMD to achieve state-of-the-art performance. Distinct from conventional machine learning methods that predict key quantities in NAMD, N$^2$AMD computes these quantities directly with a deep neural Hamiltonian, ensuring supreme accuracy, efficiency, and consistency. Furthermore, N$^2$AMD demonstrates excellent generalizability and enables seamless integration with advanced NAMD techniques and infrastructures. Taking several extensively investigated semiconductors as the prototypical system, we successfully simulate carrier recombination in both pristine and defective systems at large scales where conventional NAMD often significantly underestimates or even qualitatively incorrectly predicts lifetimes. This framework not only boosts the efficiency and precision of NAMD simulations but also opens new avenues to advance materials research., Comment: 25 pages, 7 figures

Utilizations of silicon-based luminescent devices are restricted by the indirect-gap nature of diamond silicon. In this study, the high-throughput method is employed to expedite discoveries of direct-gap silicon crystals. The machine learning (ML) potential is utilized to construct a dataset comprising 2637 silicon allotropes, which is subsequently screened using an ML Hamiltonian model and density functional theory calculations, resulting in identification of 47 direct-gap Si structures. We calculate transition dipole moments (TDM), energies, and phonon bandstructures of these structures to validate their performance. Additionally, we recalculate bandgaps of these structures employing the HSE06 functional. 22 silicon allotropes are identified as potential photovoltaic materials. Among them, the energy per atom of Si22-Pm, which has a direct bandgap of 1.27 eV, is 0.026 eV/atom higher than diamond silicon. Si18-C2/m, which has a direct bandgap of 0.796 eV, exhibits the highest TDM among identified structures. Si16-P21/c, which has a direct bandgap of 0.907 eV, has the mass density of 2.316 g/cm3, which is the highest among identified structures and higher than that of diamond silicon. The structure Si12-P1, which possesses a direct bandgap of 1.69 eV, exhibits the highest spectroscopic limited maximum efficiency (SLME) among identified structures at 32.28%, surpassing that of diamond silicon. This study offers insights into properties of silicon crystals while presenting a systematic high-throughput method for material discovery.

This paper introduces ReflectSumm, a novel summarization dataset specifically designed for summarizing students' reflective writing. The goal of ReflectSumm is to facilitate developing and evaluating novel summarization techniques tailored to real-world scenarios with little training data, %practical tasks with potential implications in the opinion summarization domain in general and the educational domain in particular. The dataset encompasses a diverse range of summarization tasks and includes comprehensive metadata, enabling the exploration of various research questions and supporting different applications. To showcase its utility, we conducted extensive evaluations using multiple state-of-the-art baselines. The results provide benchmarks for facilitating further research in this area., Comment: LREC-COLING 2024 camera ready; code and dataset are available at https://github.com/EngSalem/ReflectSUMM

While density functional theory (DFT) serves as a prevalent computational approach in electronic structure calculations, its computational demands and scalability limitations persist. Recently, leveraging neural networks to parameterize the Kohn-Sham DFT Hamiltonian has emerged as a promising avenue for accelerating electronic structure computations. Despite advancements, challenges such as the necessity for computing extensive DFT training data to explore each new system and the complexity of establishing accurate ML models for multi-elemental materials still exist. Addressing these hurdles, this study introduces a universal electronic Hamiltonian model trained on Hamiltonian matrices obtained from first-principles DFT calculations of nearly all crystal structures on the Materials Project. We demonstrate its generality in predicting electronic structures across the whole periodic table, including complex multi-elemental systems, solid-state electrolytes, Moir\'e twisted bilayer heterostructure, and metal-organic frameworks (MOFs). Moreover, we utilize the universal model to conduct high-throughput calculations of electronic structures for crystals in GeNOME datasets, identifying 3,940 crystals with direct band gaps and 5,109 crystals with flat bands. By offering a reliable efficient framework for computing electronic properties, this universal Hamiltonian model lays the groundwork for advancements in diverse fields, such as easily providing a huge data set of electronic structures and also making the materials design across the whole periodic table possible., Comment: 20 pages, 9 figures

To extend the antenna design on printed circuit boards (PCBs) for more engineers of interest, we propose a simple method that models PCB antennas with a few basic components. By taking two separate steps to decide their geometric dimensions and positions, antenna prototypes can be facilitated with no experience required. Random sampling statistics relate to the quality of dimensions are used in selecting among dimension candidates. A novel image-based classifier using a convolutional neural network (CNN) is introduced to further determine the positions of these fixed-dimension components. Two examples from wearable products have been chosen to examine the entire workflow. Their final designs are realistic and their performance metrics are not inferior to the ones designed by experienced engineers., Comment: 13 pages, 18 figures

This paper presents an overview of the ImageArg shared task, the first multimodal Argument Mining shared task co-located with the 10th Workshop on Argument Mining at EMNLP 2023. The shared task comprises two classification subtasks - (1) Subtask-A: Argument Stance Classification; (2) Subtask-B: Image Persuasiveness Classification. The former determines the stance of a tweet containing an image and a piece of text toward a controversial topic (e.g., gun control and abortion). The latter determines whether the image makes the tweet text more persuasive. The shared task received 31 submissions for Subtask-A and 21 submissions for Subtask-B from 9 different teams across 6 countries. The top submission in Subtask-A achieved an F1-score of 0.8647 while the best submission in Subtask-B achieved an F1-score of 0.5561., Comment: In The 10th Argument Mining Workshop, held in conjunction with The Conference on Empirical Methods in Natural Language Processing (EMNLP), December 2023

We propose an approach for the structure controllable summarization of long legal opinions that considers the argument structure of the document. Our approach involves using predicted argument role information to guide the model in generating coherent summaries that follow a provided structure pattern. We demonstrate the effectiveness of our approach on a dataset of legal opinions and show that it outperforms several strong baselines with respect to ROUGE, BERTScore, and structure similarity., Comment: To appear in Findings of IJCNLP-AACL 2023

Using machine learning methods, we explore different types of domain walls in the recently unveiled single-element ferroelectric, the bismuth monolayer [Nature 617, 67 (2023)]. Remarkably, our investigation reveals that the charged domain wall configuration exhibits lower energy compared to the uncharged domain wall structure. We also demonstrate that the experimentally discovered tail-to-tail domain wall maintains topological interfacial states caused by the change in the Z_2 number between ferroelectric and paraelectric states. Interestingly, due to the intrinsic built-in electric fields in asymmetry DW configurations, we find that the energy of topological interfacial states splits, resulting in an accidental band crossing at the Fermi level. Our study suggests that domain walls in two-dimensional bismuth hold potential as a promising platform for the development of ferroelectric domain wall devices., Comment: 15 pages, 4 figs

The study of the electronic properties of charged defects is crucial for our understanding of various electrical properties of materials. However, the high computational cost of density functional theory (DFT) hinders the research on large defect models. In this study, we present an E(3) equivariant graph neural network framework (HamGNN-Q), which can predict the tight-binding Hamiltonian matrices for various defect types with different charges using only one set of network parameters. By incorporating features of background charge into the element representation, HamGNN-Q enables a direct mapping from structure and background charge to the electronic Hamiltonian matrix of charged defect systems without DFT calculation. We demonstrate the model's high precision and transferability through testing on GaAs systems with various charged defect configurations. Our approach provides a practical solution for accelerating charged defect electronic structure calculations and advancing the design of materials with tailored electronic properties., Comment: 16 pages, 7 figures

Complex spin-spin interactions in magnets can often lead to magnetic superlattices with complex local magnetic arrangements, and many of the magnetic superlattices have been found to possess non-trivial topological electronic properties. Due to the huge size and complex magnetic moment arrangement of the magnetic superlattices, it is a great challenge to perform a direct DFT calculation on them. In this work, an equivariant deep learning framework is designed to accelerate the electronic calculation of magnetic systems by exploiting both the equivariant constraints of the magnetic Hamiltonian matrix and the physical rules of spin-spin interactions. This framework can bypass the costly self-consistent iterations and build a direct mapping from a magnetic configuration to the ab initio Hamiltonian matrix. After training on the magnets with random magnetic configurations, our model achieved high accuracy on the test structures outside the training set, such as spin spiral and non-collinear antiferromagnetic configurations. The trained model is also used to predict the energy bands of a skyrmion configuration of NiBrI containing thousands of atoms, showing the high efficiency of our model on large magnetic superlattices., Comment: 15 pages, 5 figures

We propose a simple approach for the abstractive summarization of long legal opinions that considers the argument structure of the document. Legal opinions often contain complex and nuanced argumentation, making it challenging to generate a concise summary that accurately captures the main points of the legal opinion. Our approach involves using argument role information to generate multiple candidate summaries, then reranking these candidates based on alignment with the document's argument structure. We demonstrate the effectiveness of our approach on a dataset of long legal opinions and show that it outperforms several strong baselines.

Electron-phonon coupling (EPC) plays an important role in many fundamental physical phenomena, but the high computational cost of the EPC matrix hinders the theoretical research on them. In this paper, an analytical formula is derived to calculate the EPC matrix in terms of the Hamiltonian and its gradient in the nonorthogonal atomic orbital bases. The recently-developed E(3) equivariant neural network is used to directly predict the Hamiltonian and its gradient needed by the formula, thus bypassing the expensive self-consistent iterations in DFT. The correctness of the proposed EPC calculation formula and the accuracy of the predicted EPC values of the network are illustrated by the tests on a water molecule and a MoS2 crystal., Comment: 11 pages, 2 figures, 2 tables

This study introduces time-reversal E(3)-equivariant neural network and SpinGNN++ framework for constructing a comprehensive interatomic potential for magnetic systems, encompassing spin-orbit coupling and noncollinear magnetic moments. SpinGNN++ integrates multitask spin equivariant neural network with explicit spin-lattice terms, including Heisenberg, Dzyaloshinskii-Moriya, Kitaev, single-ion anisotropy, and biquadratic interactions, and employs time-reversal equivariant neural network to learn high-order spin-lattice interactions using time-reversal E(3)-equivariant convolutions. To validate SpinGNN++, a complex magnetic model dataset is introduced as a benchmark and employed to demonstrate its capabilities. SpinGNN++ provides accurate descriptions of the complex spin-lattice coupling in monolayer CrI$_3$ and CrTe$_2$, achieving sub-meV errors. Importantly, it facilitates large-scale parallel spin-lattice dynamics, thereby enabling the exploration of associated properties, including the magnetic ground state and phase transition. Remarkably, SpinGNN++ identifies a new ferrimagnetic state as the ground magnetic state for monolayer CrTe2, thereby enriching its phase diagram and providing deeper insights into the distinct magnetic signals observed in various experiments., Comment: 27 pages,6 figures and 3 tables

Though many algorithms can be used to automatically summarize legal case decisions, most fail to incorporate domain knowledge about how important sentences in a legal decision relate to a representation of its document structure. For example, analysis of a legal case summarization dataset demonstrates that sentences serving different types of argumentative roles in the decision appear in different sections of the document. In this work, we propose an unsupervised graph-based ranking model that uses a reweighting algorithm to exploit properties of the document structure of legal case decisions. We also explore the impact of using different methods to compute the document structure. Results on the Canadian Legal Case Law dataset show that our proposed method outperforms several strong baselines., Comment: NLLP Workshop Camera Ready in EMNLP 2022

Strongly correlated materials often exhibit an electronic phase separation (EPS) phenomena whose domain pattern is random in nature. The ability to control the spatial arrangement of the electronic phases at microscopic scales is highly desirable for tailoring their macroscopic properties and/or designing novel electronic devices. Here we report the formation of EPS nanoscale network in a mono-atomically stacked LaMnO3/CaMnO3/PrMnO3 superlattice grown on SrTiO3 (STO) (001) substrate, which is known to have an antiferromagnetic (AFM) insulating ground state. The EPS nano-network is a consequence of an internal strain relaxation triggered by the structural domain formation of the underlying STO substrate at low temperatures. The same nanoscale network pattern can be reproduced upon temperature cycling allowing us to employ different local imaging techniques to directly compare the magnetic and transport state of a single EPS domain. Our results confirm the one-to-one correspondence between ferromagnetic (AFM) to metallic (insulating) state in manganite. It also represents a significant step in a paradigm shift from passively characterizing EPS in strongly correlated systems to actively engaging in its manipulation.

Using the message-passing mechanism in machine learning (ML) instead of self-consistent iterations to directly build the mapping from structures to electronic Hamiltonian matrices will greatly improve the efficiency of density functional theory (DFT) calculations. In this work, we proposed a general analytic Hamiltonian representation in an E(3) equivariant framework, which can fit the ab initio Hamiltonian of molecules and solids by a complete data-driven method and are equivariant under rotation, space inversion, and time reversal operations. Our model reached state-of-the-art precision in the benchmark test and accurately predicted the electronic Hamiltonian matrices and related properties of various periodic and aperiodic systems, showing high transferability and generalization ability. This framework provides a general transferable model that can be used to accelerate the electronic structure calculations on different large systems with the same network weights trained on small structures., Comment: 33 pages, 6 figures