Search

Nearly all cool, evolved stars are solar-like oscillators, and fundamental stellar properties can be inferred from these oscillations with asteroseismology. Scaling relations are commonly used to relate global asteroseismic properties, the frequency of maximum power $\nu_{max}$ and the large frequency separation $\Delta \nu$, to stellar properties. Mass, radius, and age can then be inferred with the addition of stellar spectroscopy. There is excellent agreement between seismic radii and fundamental data on the lower red giant branch and red clump. However, the scaling relations appear to breakdown in luminous red giant stars. We attempt to constrain the contributions of the asteroseismic parameters to the observed breakdown. We test the $\nu_{max}$ and $\Delta \nu$ scaling relations separately, by using stars of known mass and radius in star clusters and the Milky Way's high-$\alpha$ sequence. We find evidence that the $\Delta \nu$-scaling relation contributes to the observed breakdown in luminous giants more than the $\nu_{max}$ relation. We test different methods of mapping the observed $\Delta \nu$ to the mean density via a correction factor, $F_{\Delta \nu}$ and find a $\approx 1 - 3\%$ difference in the radii in the luminous giant regime depending on the technique used to measure $F_{\Delta \nu}$. The differences between the radii inferred by these two techniques are too small on the luminous giant branch to account for the inflated seismic radii observed in evolved giant stars. Finally, we find that the $F_{\Delta \nu}$ correction is insensitive to the adopted mixing length, chosen by calibrating the models to observations of $T_{eff}$., Comment: 28 pages, 15 figures, and 5 tables in text. 4 additional pages in the appendix. Submitted to ApJ

Computing ground-state properties of molecules is a promising application for quantum computers operating in concert with classical high-performance computing resources. Quantum embedding methods are a family of algorithms particularly suited to these computational platforms: they combine high-level calculations on active regions of a molecule with low-level calculations on the surrounding environment, thereby avoiding expensive high-level full-molecule calculations and allowing to distribute computational cost across multiple and heterogeneous computing units. Here, we present the first density matrix embedding theory (DMET) simulations performed in combination with the sample-based quantum diagonalization (SQD) method. We employ the DMET-SQD formalism to compute the ground-state energy of a ring of 18 hydrogen atoms, and the relative energies of the chair, half-chair, twist-boat, and boat conformers of cyclohexane. The full-molecule 41- and 89-qubit simulations are decomposed into 27- and 32-qubit active-region simulations, that we carry out on the ibm_cleveland device, obtaining results in agreement with reference classical methods. Our DMET-SQD calculations mark a tangible progress in the size of active regions that can be accurately tackled by near-term quantum computers, and are an early demonstration of the potential for quantum-centric simulations to accurately treat the electronic structure of large molecules, with the ultimate goal of tackling systems such as peptides and proteins.

We present the first quantum-centric simulations of noncovalent interactions using a supramolecular approach. We simulate the potential energy surfaces (PES) of the water and methane dimers, featuring hydrophilic and hydrophobic interactions, respectively, with a sample-based quantum diagonalization (SQD) approach. Our simulations on quantum processors, using 27- and 36-qubit circuits, are in remarkable agreement with classical methods, deviating from complete active space configuration interaction (CASCI) and coupled-cluster singles, doubles, and perturbative triples (CCSD(T)) within 1 kcal/mol in the equilibrium regions of the PES. Finally, we test the capacity limits of the quantum methods for capturing hydrophobic interactions with an experiment on 54 qubits. These results mark significant progress in the application of quantum computing to chemical problems, paving the way for more accurate modeling of noncovalent interactions in complex systems critical to the biological, chemical and pharmaceutical sciences.

We present Benchpress, a benchmarking suite for evaluating the performance and range of functionality of multiple quantum computing software development kits. This suite consists of a collection of over $1000$ tests measuring key performance metrics for a wide variety of operations on quantum circuits comprised of up to $930$ qubits and $\mathcal{O}(10^{6})$ two-qubit gates, as well as an execution framework for running the tests over multiple quantum software packages in a unified manner. We give a detailed overview of the benchmark suite, its methodology, and generate representative results over seven different quantum software packages. The flexibility of the Benchpress framework allows for benchmarking that not only keeps pace with quantum hardware improvements but can preemptively gauge the quantum circuit processing costs of future device architectures. Being open-source, Benchpress ensures the transparency and verification of performance claims., Comment: 11 pages, 4 figures, 2 tables

Physics, like many scientific disciplines, has long struggled with attracting and retaining a diverse population and fostering inclusivity. While there have been improvements in addressing equity issues within physics, significant challenges remain. Faculty members play a crucial role as change agents in promoting cultural transformation within academic environments. Empathy, a fundamental component of effective teaching, mentoring, and collegiality, is essential for fostering a student-centered and holistic approach in academia. However, understanding how empathy functions within the specific context of physics, including its interaction with power dynamics and other contextual factors, remains underexplored.This study presents a theoretical model of empathy development among physics faculty as they engage with students and colleagues. Conducted at a private R2 institution, the study involved four rounds of interviews from summer 2023 to spring 2024. The initial two rounds (summer 2023) included eight participants, followed by 19 participants in fall 2023, and nine participants chose to follow up with us in spring 2024. All participants were physics faculty members, either in teaching or tenure-track positions. The developed model builds on previous research by introducing new complexities in the understanding of empathy. It identifies key mediators, including reflective witnessing, personal experiences, and empathetic concern, as well as moderators such as individual experience, emotions, motivation, values, and situational information. The model delineates both cognitive and affective pathways of empathy, providing a nuanced framework for understanding how empathy develops and influences faculty interactions in the physics discipline.

Dark matter (DM) bars that shadow stellar bars have been previously shown to form in idealized simulations of isolated disk galaxies. Here, we show that DM bars commonly occur in barred disk galaxies in the TNG50 cosmological simulation suite, but do not appear in unbarred disk galaxies. Consistent with isolated simulations, DM bars are typically shorter than their stellar counterparts and are $75\%$ weaker as measured by the Fourier $A_2$ moment. DM bars dominate the shape of the inner halo potential and are easily identified in the time series of quadrupolar coefficients. We present two novel methods for measuring the bar pattern speed using these coefficients, and use them to make a measurement of the pattern speed and rotation axis orientation for one sample galaxy located in one of the high-time resolution subboxes of TNG50. The stellar and dark bars in this galaxy remain co-aligned throughout the last 8 Gyr and have identical pattern speeds throughout. Both the pattern speed and rotation axis orientation of the bars evolve considerably over the last 8 Gyr, consistent with torques on the bars due to dynamical friction and gas accretion. While the bar pattern speed generally decreases over the time course, it is seen to increase after mergers. Rather than remaining static in time, the rotation axis displays both precession and nutation possibly caused by torques outside the plane of rotation. We find that the shape of the stellar and DM mass distributions are tightly correlated with the bar pattern speed., Comment: 11 pages, 4 figures

Stellar streams that emerge from globular clusters (GCs) are thin stellar structures spread along the orbits of progenitor clusters. Numerical modeling of these streams is essential for understanding their interaction with the host galaxy's mass distribution. Traditional methods are either computationally expensive or oversimplified, motivating us to develop a fast and accurate approach using a particle spray algorithm. By conducting a series of N-body simulations of GCs orbiting a host galaxy, we find that the position and velocity distributions of newly-escaped stream particles are consistent across various GC masses and orbital parameters. Based on these distributions, we develop a new algorithm that avoids computing the detailed internal cluster dynamics by directly drawing tracer particles from these distributions. This algorithm correctly reproduces the action space distribution of stream particles and achieves a 10% accuracy in stream morphology and velocities compared to N-body simulations. To facilitate broader use, we have implemented this algorithm in galactic dynamics codes agama, gala, galax, and galpy., Comment: 18 pages, 11 figures, accepted for publication in ApJS. Online notebooks providing instructions on this algorithm are available at https://github.com/ybillchen/particle_spray

Retrieval Augmented Generation (RAG) is widely employed to ground responses to queries on domain-specific documents. But do RAG implementations leave out important information when answering queries that need an integrated analysis of information (e.g., Tell me good news in the stock market today.)? To address these concerns, RAG developers need to annotate information retrieval (IR) data for their domain of interest, which is challenging because (1) domain-specific queries usually need nuanced definitions of relevance beyond shallow semantic relevance; and (2) human or GPT-4 annotation is costly and cannot cover all (query, document) pairs (i.e., annotation selection bias), thus harming the effectiveness in evaluating IR recall. To address these challenges, we propose DIRAS (Domain-specific Information Retrieval Annotation with Scalability), a manual-annotation-free schema that fine-tunes open-sourced LLMs to consider nuanced relevance definition and annotate (partial) relevance labels with calibrated relevance scores. Extensive evaluation shows that DIRAS enables smaller (8B) LLMs to achieve GPT-4-level performance on annotating and ranking unseen (query, document) pairs, and is helpful for real-world RAG development. All code, LLM generations, and human annotations can be found in \url{https://github.com/EdisonNi-hku/DIRAS}.

Large language models such as ChatGPT exhibit striking political biases. If users query them about political information, they often take a normative stance. To overcome this, we align LLMs with diverse political viewpoints from 100,000 comments written by candidates running for national parliament in Switzerland. Models aligned with this data can generate more accurate political viewpoints from Swiss parties, compared to commercial models such as ChatGPT. We also propose a procedure to generate balanced overviews summarizing multiple viewpoints using such models. The replication package contains all code and data., Comment: accepted at EMNLP 2024 main as a short paper

We report the synthesis, structural characterization, and investigation of electrical transport, magnetic and specific heat properties of bulk semiconducting layered material Ta$_2$Ni$_3$Te$_5$. Ta$_2$Ni$_3$Te$_5$ crystallizes in the centrosymmetric orthorhombic structure with space group Pnma. Temperature-dependent resistivity shows a transition from semiconducting to metallic nature below 7 K. Low-temperature magnetotransport studies show large magnetoresistance with the signature of weak anti-localization (WAL) effect. The magnetoconductivity data has been used to explore the origin of the WAL effect, extract relevant parameters, and study their variation with temperature. The presence of significant electron-phonon interaction is evident from the MR vs. B/R plot (Kohler's plot). Isothermal field-dependent magnetization studies show Berry paramagnetism as the signature for spin-orbit coupling-induced spin-momentum locking. Observation of WAL effect and spin-momentum locking phenomenon demonstrate Ta$_2$Ni$_3$Te$_5$ as a promising low dimensional material for quantum spin Hall insulator-based applications., Comment: 10 pages

There is a growing body of work on learning from human feedback to align various aspects of machine learning systems with human values and preferences. We consider the setting of fairness in content moderation, in which human feedback is used to determine how two comments -- referencing different sensitive attribute groups -- should be treated in comparison to one another. With a novel dataset collected from Prolific and MTurk, we find significant gaps in fairness preferences depending on the race, age, political stance, educational level, and LGBTQ+ identity of annotators. We also demonstrate that demographics mentioned in text have a strong influence on how users perceive individual fairness in moderation. Further, we find that differences also exist in downstream classifiers trained to predict human preferences. Finally, we observe that an ensemble, giving equal weight to classifiers trained on annotations from different demographics, performs better for different demographic intersections; compared to a single classifier that gives equal weight to each annotation., Comment: To appear in the Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics, ACL 2024

Continual learning with deep neural networks presents challenges distinct from both the fixed-dataset and convex continual learning regimes. One such challenge is plasticity loss, wherein a neural network trained in an online fashion displays a degraded ability to fit new tasks. This problem has been extensively studied in both supervised learning and off-policy reinforcement learning (RL), where a number of remedies have been proposed. Still, plasticity loss has received less attention in the on-policy deep RL setting. Here we perform an extensive set of experiments examining plasticity loss and a variety of mitigation methods in on-policy deep RL. We demonstrate that plasticity loss is pervasive under domain shift in this regime, and that a number of methods developed to resolve it in other settings fail, sometimes even performing worse than applying no intervention at all. In contrast, we find that a class of ``regenerative'' methods are able to consistently mitigate plasticity loss in a variety of contexts, including in gridworld tasks and more challenging environments like Montezuma's Revenge and ProcGen.

We extend the computations in our prior work to find the cohomology in degree five of a congruence subgroup Gamma of SL_4(Z) with coefficients in Sym^g(K^4), twisted by a nebentype character eta, along with the action of the Hecke algebra. This is the top cuspidal degree. In this paper we take K to be a finite field of large characteristic, as a proxy for the complex numbers. For each Hecke eigenclass found, we produce the unique Galois representation that appears to be attached to it. The computations require modifications to our previous algorithms to accommodate the fact that the coefficients are not one-dimensional., Comment: arXiv admin note: text overlap with arXiv:1806.08707

We prove that the direct sum of all homology groups of the integral general linear groups with Steinberg module coefficients form a commutative Hopf algebra, in particular a free graded commutative algebra. We use this to construct new infinite families of unstable cohomology classes of $SL_n(\mathbb Z)$., Comment: 11 pages

The ability to emotionally or intellectually understand another person's thoughts and feelings-empathy-can foster critical connections that facilitate learning and collaboration. We present a case study of physics faculty that examines their experiences empathizing with students, both in and outside of the classroom. We expand on frameworks for understanding the empathy process by identifying key mediating factors, and note various barriers that faculty express as preventing them from taking empathetic action. Our analysis unpacks the mechanisms of communication and contextual information, which play key roles in the empathetic process, with implications for programs that rely on empathy to develop more inclusive STEM academic spaces., Comment: ISLS owns the copyright of this paper as it will appear in the upcoming 2024 Conference Proceedings. We have obtained permission to share this preprint

This study investigates how faculty acquire contextual information about students, examining mechanisms and motivations used when sharing their identity to facilitate empathy. Empathy is "the ability and tendency to share and understand others' internal state" (Zaki and Ochsner, 2012), and is a critical factor in both motivating faculty to enact large scale change and take immediate, smaller actions. This study explores the impact identity sharing has on obtaining contextual information that motivates empathetic action. Nineteen semi structured interviews with physics faculty explored participant identities and interactions across various contexts. Employing emergent thematic coding, we crafted four personas around faculty sharing, teaching values, and student reciprocity. Brooke, the Trust Builder, prioritizes creating an environment of trust by openly discussing their identity, aiming to foster student openness. Nour, the Identity Navigator, shares personal experiences to assist others in navigating their own identities, acknowledging the challenges of college years. Brooke and Nour had more students approaching them with personal issues, indicating a correlation between faculty identity sharing and student openness. Casey, the Cautious Sharer, expresses concerns about potential alienation or backlash, approaching personal sharing with caution. Wray, adopting a Walled Off approach, separates personal and professional life due to past negative experiences or a belief in the importance of that division. Among the faculty interviewed, 15 who were explicitly open about their identities reported that students approached them with personal issues. This study outlines mechanisms influencing when and what faculty share about themselves in different contexts. Our findings underscore the significance of fostering dialogue as the initial step in empathy development.

This report describes and synthesizes implementation determinants and strategies to promote uptake and sustained implementation of comprehensive school mental health and trauma-sensitive practices in schools. The present synthesis of studies and reports emphasizes use of relevant and efficient practices with empirical support at the state, district, and school levels to promote comprehensive school mental health and trauma-sensitive practices.

Youth mental health concerns are on the rise and can have serious implications for student well-being and success (Clayborne et al., 2019). Consequently, more educators and mental health professionals are identifying mental health as an important student need (American Academy of Pediatrics, 2021). This report summarizes the key principles underlying comprehensive school mental health and trauma sensitive schools. The present synthesis of studies and reports emphasizes core features, key principles, and associated core components. [This report was co-produced by the Wisconsin Evaluation Collaborative and originally prepared for the Wisconsin Department of Public Instruction (DPI).]

On-chip photonic processors for neural networks have potential benefits in both speed and energy efficiency but have not yet reached the scale at which they can outperform electronic processors. The dominant paradigm for designing on-chip photonics is to make networks of relatively bulky discrete components connected by one-dimensional waveguides. A far more compact alternative is to avoid explicitly defining any components and instead sculpt the continuous substrate of the photonic processor to directly perform the computation using waves freely propagating in two dimensions. We propose and demonstrate a device whose refractive index as a function of space, $n(x,z)$, can be rapidly reprogrammed, allowing arbitrary control over the wave propagation in the device. Our device, a 2D-programmable waveguide, combines photoconductive gain with the electro-optic effect to achieve massively parallel modulation of the refractive index of a slab waveguide, with an index modulation depth of $10^{-3}$ and approximately $10^4$ programmable degrees of freedom. We used a prototype device with a functional area of $12\,\text{mm}^2$ to perform neural-network inference with up to 49-dimensional input vectors in a single pass, achieving 96% accuracy on vowel classification and 86% accuracy on $7 \times 7$-pixel MNIST handwritten-digit classification. This is a scale beyond that of previous photonic chips relying on discrete components, illustrating the benefit of the continuous-waves paradigm. In principle, with large enough chip area, the reprogrammability of the device's refractive index distribution enables the reconfigurable realization of any passive, linear photonic circuit or device. This promises the development of more compact and versatile photonic systems for a wide range of applications, including optical processing, smart sensing, spectroscopy, and optical communications.

Denote the virtual cohomological dimension of SL_n(Z) by t=n(n-1)/2. Let St denote the Steinberg module of SL_n(Q) tensored with Q. Let Sh_* denote the sharbly resolution of the Steinberg module St. By Borel-Serre duality, the one-dimensional Q-vector space H^0(SL_n(Z), Q) is isomorphic to H_t(SL_n(Z),St). We find an explicit generator of H_t(SL_n(Z),St) in terms of sharbly cycles and cosharbly cocycles. These methods may extend to other degrees of cohomology of SL_n(Z).

With the rise of generative AI, automated fact-checking methods to combat misinformation are becoming more and more important. However, factual claim detection, the first step in a fact-checking pipeline, suffers from two key issues that limit its scalability and generalizability: (1) inconsistency in definitions of the task and what a claim is, and (2) the high cost of manual annotation. To address (1), we review the definitions in related work and propose a unifying definition of factual claims that focuses on verifiability. To address (2), we introduce AFaCTA (Automatic Factual Claim deTection Annotator), a novel framework that assists in the annotation of factual claims with the help of large language models (LLMs). AFaCTA calibrates its annotation confidence with consistency along three predefined reasoning paths. Extensive evaluation and experiments in the domain of political speech reveal that AFaCTA can efficiently assist experts in annotating factual claims and training high-quality classifiers, and can work with or without expert supervision. Our analyses also result in PoliClaim, a comprehensive claim detection dataset spanning diverse political topics., Comment: ACL2024 Main Conference

Advances towards more faithful and traceable answers of Large Language Models (LLMs) are crucial for various research and practical endeavors. One avenue in reaching this goal is basing the answers on reliable sources. However, this Evidence-Based QA has proven to work insufficiently with LLMs in terms of citing the correct sources (source quality) and truthfully representing the information within sources (answer attributability). In this work, we systematically investigate how to robustly fine-tune LLMs for better source quality and answer attributability. Specifically, we introduce a data generation pipeline with automated data quality filters, which can synthesize diversified high-quality training and testing data at scale. We further introduce four test sets to benchmark the robustness of fine-tuned specialist models. Extensive evaluation shows that fine-tuning on synthetic data improves performance on both in- and out-of-distribution. Furthermore, we show that data quality, which can be drastically improved by proposed quality filters, matters more than quantity in improving Evidence-Based QA.