Search

In this study, we highlight the differences in classroom-, teacher-, and school-level factors in 8th and 9th grade algebra experiences along socioeconomic and racial/ethnic lines using nationally representative survey data from the American Mathematics Educator Study. Several takeaways emerge from our analysis. First, we show that highest-poverty schools (i.e., schools in the top poverty quartile) are significantly less likely to offer algebra in 8th grade unconditionally (i.e., without needing to meet certain conditions) for all students or to offer algebra at all compared to lowest-poverty schools (i.e., schools in the lowest poverty quartile). Second, we find significant differences in which factors (e.g., parent requests, teacher referrals) are considered when placing students in advanced math courses in 8th and 9th grade that may affect the access of students from disadvantaged backgrounds to these courses or to more advanced pathways. Third, we show significant differences in 8th and 9th grade math teacher qualifications and classroom activities in math courses, with teachers in highest-poverty schools being significantly more likely to have received alternative credentials, less likely to have completed student-teaching during their preparation program, and less likely to have completed their state's licensure requirements for math. 8th and 9th grade math teachers in highest-poverty schools were also more likely to report that they spend more than half of their instruction time addressing math topics below grade level or addressing disciplinary issues. Mostly similar, albeit weaker, patterns emerge when we examine discrepancies along school racial/ethnic composition. Offering 8th grade algebra in high-poverty school settings (or making it available to more or all students) could help close socioeconomic gaps in algebra enrollment in 8th grade and grant more equitable access to advanced math coursework in the long-run. That said, focusing on the provision of 8th grade algebra alone will likely not remedy the opportunity gaps in access to (and completion of) advanced math courses in high school since our findings suggest that highest-poverty high schools are also significantly less likely to offer college credit-bearing math courses. Further, our findings suggest that increasing the provision of algebra in 8th grade may present three challenges: (1) staffing these courses with qualified teachers; (2) providing strong supports for students who struggle with algebra; and, relatedly, (3) making algebra placement decisions that minimize failure and maximize success for the greatest number of students. Taken together, our findings demonstrate systemic inequities across racial/ethnic and socioeconomic lines in terms of access to, and experiences in, 8th and 9th grade math courses.

Math achievement plays a pivotal role in students' success in school and beyond. Algebra, in particular, serves as a crucial gateway: it increases the chances a student takes advanced math in high school, which has implications for their college and career prospects. In 2023 and 2024, the American Mathematics Educator Study (AMES) surveyed teachers and principals nationwide to learn more about how students access and experience algebra in 8th and 9th grade. A key finding from the surveys: schools serving mostly low-income students are far less likely to offer algebra in 8th grade compared to those serving more affluent students. The surveys also identified other issues that can hinder equitable access to algebra and math education, including: (1) uneven access to qualified math teachers; (2) math instruction that disproportionately focuses on remedial content in disadvantaged communities; (3) limited access to advanced math courses beyond algebra in high-poverty schools; and (4) different approaches to algebra eligibility and achievement grouping (e.g., tracking) that may perpetuate inequity. These systemic inequities--in both opportunities and experiences--suggest that making algebra more available is a necessary but insufficient step towards closing gaps in access to advanced math education. The brief is based on a working paper which examines how America's students access and experience algebra in 8th and 9th grade.

TikTok is a major force among social media platforms with over a billion monthly active users worldwide and 170 million in the United States. The platform's status as a key news source, particularly among younger demographics, raises concerns about its potential influence on politics in the U.S. and globally. Despite these concerns, there is scant research investigating TikTok's recommendation algorithm for political biases. We fill this gap by conducting 323 independent algorithmic audit experiments testing partisan content recommendations in the lead-up to the 2024 U.S. presidential elections. Specifically, we create hundreds of "sock puppet" TikTok accounts in Texas, New York, and Georgia, seeding them with varying partisan content and collecting algorithmic content recommendations for each of them. Collectively, these accounts viewed ~394,000 videos from April 30th to November 11th, 2024, which we label for political and partisan content. Our analysis reveals significant asymmetries in content distribution: Republican-seeded accounts received ~11.8% more party-aligned recommendations compared to their Democratic-seeded counterparts, and Democratic-seeded accounts were exposed to ~7.5% more opposite-party recommendations on average. These asymmetries exist across all three states and persist when accounting for video- and channel-level engagement metrics such as likes, views, shares, comments, and followers, and are driven primarily by negative partisanship content. Our findings provide insights into the inner workings of TikTok's recommendation algorithm during a critical election period, raising fundamental questions about platform neutrality., Comment: 19 pages, 5 Figures, 1 Table

We present Explainable XR, an end-to-end framework for analyzing user behavior in diverse eXtended Reality (XR) environments by leveraging Large Language Models (LLMs) for data interpretation assistance. Existing XR user analytics frameworks face challenges in handling cross-virtuality - AR, VR, MR - transitions, multi-user collaborative application scenarios, and the complexity of multimodal data. Explainable XR addresses these challenges by providing a virtuality-agnostic solution for the collection, analysis, and visualization of immersive sessions. We propose three main components in our framework: (1) A novel user data recording schema, called User Action Descriptor (UAD), that can capture the users' multimodal actions, along with their intents and the contexts; (2) a platform-agnostic XR session recorder, and (3) a visual analytics interface that offers LLM-assisted insights tailored to the analysts' perspectives, facilitating the exploration and analysis of the recorded XR session data. We demonstrate the versatility of Explainable XR by demonstrating five use-case scenarios, in both individual and collaborative XR applications across virtualities. Our technical evaluation and user studies show that Explainable XR provides a highly usable analytics solution for understanding user actions and delivering multifaceted, actionable insights into user behaviors in immersive environments., Comment: 11 pages, 8 figures. This is the author's version of the article that has been accepted for publication in IEEE Transactions on Visualization and Computer Graphics

New media art builds on top of rich software stacks. Blending multiple media such as code, light or sound , new media artists integrate various types of software to draw, animate, control or synchronize different parts of an artwork. Yet, the artworks rarely credit software and all the developers involved. In this work, we present Myriad People, an original dataset of open source projects and their contributors, which span various software layers used in new media art installations. To collect this dataset, we released an open call for artists and eventually curated 9 artworks, which use a variety of software and media. In October 2024, we organized a collective exhibition in Stockholm, entitled Myriad, which showcased the 9 artworks. The Myriad People dataset includes the 124 open source projects used in one or more of the Myriad's artworks, as well as all the contributors to these projects. In this paper, we present the dataset, as well as the possible usages of this dataset for software and art research., Comment: 5 pages, 4 figures

This paper presents a comprehensive evaluation of GitHub Copilot's deployment and impact on developer productivity at Zoominfo, a leading Go-To-Market (GTM) Intelligence Platform. We describe our systematic four-phase approach to evaluating and deploying GitHub Copilot across our engineering organization, involving over 400 developers. Our analysis combines both quantitative metrics, focusing on acceptance rates of suggestions given by GitHub Copilot and qualitative feedback given by developers through developer satisfaction surveys. The results show an average acceptance rate of 33% for suggestions and 20% for lines of code, with high developer satisfaction scores of 72%. We also discuss language-specific performance variations, limitations, and lessons learned from this medium-scale enterprise deployment. Our findings contribute to the growing body of knowledge about AI-assisted software development in enterprise settings., Comment: 25 pages, 11 figures

We pretrain METAGENE-1, a 7-billion-parameter autoregressive transformer model, which we refer to as a metagenomic foundation model, on a novel corpus of diverse metagenomic DNA and RNA sequences comprising over 1.5 trillion base pairs. This dataset is sourced from a large collection of human wastewater samples, processed and sequenced using deep metagenomic (next-generation) sequencing methods. Unlike genomic models that focus on individual genomes or curated sets of specific species, the aim of METAGENE-1 is to capture the full distribution of genomic information present within this wastewater, to aid in tasks relevant to pandemic monitoring and pathogen detection. We carry out byte-pair encoding (BPE) tokenization on our dataset, tailored for metagenomic sequences, and then pretrain our model. In this paper, we first detail the pretraining dataset, tokenization strategy, and model architecture, highlighting the considerations and design choices that enable the effective modeling of metagenomic data. We then show results of pretraining this model on our metagenomic dataset, providing details about our losses, system metrics, and training stability over the course of pretraining. Finally, we demonstrate the performance of METAGENE-1, which achieves state-of-the-art results on a set of genomic benchmarks and new evaluations focused on human-pathogen detection and genomic sequence embedding, showcasing its potential for public health applications in pandemic monitoring, biosurveillance, and early detection of emerging health threats.

Many-body fermionic systems can be simulated in a hardware-efficient manner using a fermionic quantum processor. Neutral atoms trapped in optical potentials can realize such processors, where non-local fermionic statistics are guaranteed at the hardware level. Implementing quantum error correction in this setup is however challenging, due to the atom-number superselection present in atomic systems, that is, the impossibility of creating coherent superpositions of different particle numbers. In this work, we overcome this constraint and present a blueprint for an error-corrected fermionic quantum computer that can be implemented using current experimental capabilities. To achieve this, we first consider an ancillary set of fermionic modes and design a fermionic reference, which we then use to construct superpositions of different numbers of referenced fermions. This allows us to build logical fermionic modes that can be error corrected using standard atomic operations. Here, we focus on phase errors, which we expect to be a dominant source of errors in neutral-atom quantum processors. We then construct logical fermionic gates, and show their implementation for the logical particle-number conserving processes relevant for quantum simulation. Finally, our protocol is illustrated using a minimal fermionic circuit, where it leads to a quadratic suppression of the logical error rate.

Cholesky linear solvers are a critical bottleneck in challenging applications within computer graphics and scientific computing. These applications include but are not limited to elastodynamic barrier methods such as Incremental Potential Contact (IPC), and geometric operations such as remeshing and morphology. In these contexts, the sparsity patterns of the linear systems frequently change across successive calls to the Cholesky solver, necessitating repeated symbolic analyses that dominate the overall solver runtime. To address this bottleneck, we evaluate our method on over 150,000 linear systems generated from diverse nonlinear problems with dynamic sparsity changes in Incremental Potential Contact (IPC) and patch remeshing on a wide range of triangular meshes of various sizes. Our analysis using three leading sparse Cholesky libraries, Intel MKL Pardiso, SuiteSparse CHOLMOD, and Apple Accelerate, reveals that the primary performance constraint lies in the symbolic re-ordering phase of the solver. Recognizing this, we introduce Parth, an innovative re-ordering method designed to update ordering vectors only where local connectivity changes occur adaptively. Parth employs a novel hierarchical graph decomposition algorithm to break down the dual graph of the input matrix into fine-grained subgraphs, facilitating the selective reuse of fill-reducing orderings when sparsity patterns exhibit temporal coherence. Our extensive evaluation demonstrates that Parth achieves up to a 255x and 13x speedup in fill-reducing ordering for our IPC and remeshing benchmark and a 6.85x and 10.7x acceleration in symbolic analysis. These enhancements translate to up to 2.95x and 5.89x reduction in overall solver runtime. Additionally, Parth's integration requires only three lines of code, resulting in significant computational savings without the requirement of changes to the computational stack.

Even with various regulations in place across countries and social media platforms (Government of India, 2021; European Parliament and Council of the European Union, 2022, digital abusive speech remains a significant issue. One potential approach to address this challenge is automatic text detoxification, a text style transfer (TST) approach that transforms toxic language into a more neutral or non-toxic form. To date, the availability of parallel corpora for the text detoxification task (Logachevavet al., 2022; Atwell et al., 2022; Dementievavet al., 2024a) has proven to be crucial for state-of-the-art approaches. With this work, we extend parallel text detoxification corpus to new languages -- German, Chinese, Arabic, Hindi, and Amharic -- testing in the extensive multilingual setup TST baselines. Next, we conduct the first of its kind an automated, explainable analysis of the descriptive features of both toxic and non-toxic sentences, diving deeply into the nuances, similarities, and differences of toxicity and detoxification across 9 languages. Finally, based on the obtained insights, we experiment with a novel text detoxification method inspired by the Chain-of-Thoughts reasoning approach, enhancing the prompting process through clustering on relevant descriptive attributes., Comment: COLING 2025, main conference, long

We present an optical tweezer array of $^{87}$Rb atoms housed in an cryogenic environment that successfully combines a 4 K cryopumping surface, a <50 K cold box surrounding the atoms, and a room-temperature high-numerical-aperture objective lens. We demonstrate a 3000 s atom trap lifetime, which enables us to optimize and measure losses at the $10^{-4}$ level that arise during imaging and cooling, which are important to array rearrangement. We perform both ground-state qubit manipulation with an integrated microwave antenna and two-photon coherent Rydberg control, with the local electric field tuned to zero via integrated electrodes. We anticipate that the reduced blackbody radiation at the atoms from the cryogenic environment, combined with future electrical shielding, should decrease the rate of undesired transitions to nearby strongly-interacting Rydberg states, which cause many-body loss and impede Rydberg gates. This low-vibration, high-optical-access cryogenic platform can be used with a wide range of optically trapped atomic or molecular species for applications in quantum computing, simulation, and metrology., Comment: 19 pages, 10 figures

This technical report provides detailed information about the sample, survey instruments, and resultant data for the 2024 American Instructional Resources Surveys (AIRS) that were administered to principals and teachers in spring 2024 via RAND's American Educator Panels (AEP). The 2024 AIRS focused on the usage of, perceptions of, and supports for instructional materials used in English language arts, mathematics, and science kindergarten through grade 12 classrooms across the United States. The results are intended to inform policy and education practice related to the use of instructional resources.

This paper introduces a novel XAI approach based on near-misses analysis (NMA). This approach reveals a hierarchy of logical 'concepts' inferred from the latent decision-making process of a Neural Network (NN) without delving into its explicit structure. We examined our proposed XAI approach on different network architectures that vary in size and shape (e.g., ResNet, VGG, EfficientNet, MobileNet) on several datasets (ImageNet and CIFAR100). The results demonstrate its usability to reflect NNs latent process of concepts generation. We generated a new metric for explainability. Moreover, our experiments suggest that efficient architectures, which achieve a similar accuracy level with much less neurons may still pay the price of explainability and robustness in terms of concepts generation. We, thus, pave a promising new path for XAI research to follow.

Using the bitstring probabilities of ground states of bipartitioned ladders of Rydberg atoms, we calculate the mutual information which is a lower bound on the corresponding bipartite von Neumann quantum entanglement entropy $S^{vN}_A$. We show that in many cases, these lower bounds can be improved by removing the bitstrings with a probability lower than some value $p_{min}$ and renormalizing the remaining probabilities (filtering). Surprisingly, in some cases, as we increase $p_{min}$ the filtered mutual information tends to plateaus at values very close to $S^{vN}_A$ over some range of $p_{min}$. We consider various sizes, lattice spacings and bipartitions. Our numerical investigation suggest that the filtered mutual information obtained with samples having just a few thousand bitstrings can provide reasonably close estimates of $S^{vN}_A$. We briefly discuss practical implementations with the QuEra's Aquila device., Comment: 10 pages, 12 figures, uses revtex

The reversible semiconductor-to-metal transition of vanadium dioxide (VO2) makes VO2-based coatings a promising candidate for thermochromic smart windows, reducing the energy consumption of buildings. We report low-temperature (320 degC) depositions of thermochromic V1-x-yWxSryO2 films with a thickness of 71-73 nm onto 170-175 nm thick Y-stabilized ZrO2 layers on a 1 mm thick conventional soda-lime glass. The developed deposition technique is based on reactive high-power impulse magnetron sputtering with a pulsed O2 flow feedback control allowing us to prepare crystalline W and Sr co-doped VO2 films of the required stoichiometry without any substrate bias or post-deposition annealing. The W doping of VO2 decreases the transition temperature below 25 degC, while the Sr doping of VO2 increases the integral luminous transmittance, Tlum, significantly due to widening of the visible-range optical band gap, which is consistent with lowering of the absorption coefficient of films. We present the discussion of the effect of the Sr content in the metal sublattice of VO2 on the electronic and crystal structure of V1-x-yWxSryO2 films, and on their temperature-dependent optical and electrical properties. An optimized V0.855W0.018Sr0.127O2 film exhibits a high Tlum = 56.8% and modulation of the solar energy transmittance deltaTsol = 8.3%, which are 1.50 times and 1.28 times, respectively, higher compared with those of the V0.984W0.016O2 film. The achieved results constitute an important step toward a low-temperature synthesis of large-area thermochromic VO2-based coatings for future smart-window applications, as it is easy to further increase the Tlum and deltaTsol by >6% and >3%, respectively, using a 280 nm thick top SiO2 antireflection layer.

Coboundary expansion is a high dimensional generalization of the Cheeger constant to simplicial complexes. Originally, this notion was motivated by the fact that it implies topological expansion, but nowadays a significant part of the motivation stems from its deep connection to problems in theoretical computer science such as agreement expansion in the low soundness regime. In this paper, we prove coboundary expansion with non-Abelian coefficients for the coset complex construction of Kaufman and Oppenheim. Our proof uses a novel global argument, as opposed to the local-to-global arguments that are used to prove cosystolic expansion.

Transformer models have consistently achieved remarkable results in various domains such as natural language processing and computer vision. However, despite ongoing research efforts to better understand these models, the field still lacks a comprehensive understanding. This is particularly true for deep time series forecasting methods, where analysis and understanding work is relatively limited. Time series data, unlike image and text information, can be more challenging to interpret and analyze. To address this, we approach the problem from a manifold learning perspective, assuming that the latent representations of time series forecasting models lie next to a low-dimensional manifold. In our study, we focus on analyzing the geometric features of these latent data manifolds, including intrinsic dimension and principal curvatures. Our findings reveal that deep transformer models exhibit similar geometric behavior across layers, and these geometric features are correlated with model performance. Additionally, we observe that untrained models initially have different structures, but they rapidly converge during training. By leveraging our geometric analysis and differentiable tools, we can potentially design new and improved deep forecasting neural networks. This approach complements existing analysis studies and contributes to a better understanding of transformer models in the context of time series forecasting. Code is released at https://github.com/azencot-group/GATLM., Comment: Accepted to TMLR 2024

Large language models (LLMs) acquire beliefs about gender from training data and can therefore generate text with stereotypical gender attitudes. Prior studies have demonstrated model generations favor one gender or exhibit stereotypes about gender, but have not investigated the complex dynamics that can influence model reasoning and decision-making involving gender. We study gender equity within LLMs through a decision-making lens with a new dataset, DeMET Prompts, containing scenarios related to intimate, romantic relationships. We explore nine relationship configurations through name pairs across three name lists (men, women, neutral). We investigate equity in the context of gender roles through numerous lenses: typical and gender-neutral names, with and without model safety enhancements, same and mixed-gender relationships, and egalitarian versus traditional scenarios across various topics. While all models exhibit the same biases (women favored, then those with gender-neutral names, and lastly men), safety guardrails reduce bias. In addition, models tend to circumvent traditional male dominance stereotypes and side with 'traditionally female' individuals more often, suggesting relationships are viewed as a female domain by the models., Comment: EMNLP Findings 2024

Rydberg dressing is a powerful tool for entanglement generation in long-lived atomic states. While already employed effectively in several demonstrations, a key challenge for this technique is the collective loss triggered by blackbody-radiation-driven transitions to contaminant Rydberg states of opposite parity. We demonstrate the rapid removal of such contaminants using autoionization (AI) transitions found in alkaline-earth-like atoms. The AI is shown to be compatible with coherent operation of an array of optical clock qubits. By incorporating AI pulses into a stroboscopic Rydberg dressing (SRD) sequence, we enhance lifetimes by an order of magnitude for system sizes of up to 144 atoms, while maintaining an order of magnitude larger duty cycle than previously achieved. To highlight the utility of our approach, we use the AI-enhanced SRD protocol to improve the degree of spin-squeezing achieved during early-time dressing dynamics. These results bring Rydberg dressing lifetimes closer to fundamental limits, opening the door to previously infeasible dressing proposals., Comment: 8 pages, 4 figures + Supplementary Material 7 pages, 6 figures

We define a new family of noncommutative generalizations of cluster algebras called polygonal cluster algebras. These algebras generalize the noncommutative surfaces of Berenstein-Retakh, and are inspired by the emerging theory of $\Theta$-positivity for the groups $\mathrm{Spin}(p,q)$. They are generated by mutations of quivers which we call ST-compatible, and which encode the order of the products that appear in the exchange relations. We show that these ST-compatible quivers can be represented by tilings of surfaces by polygons, a generalization of the description of surface type cluster algebras. As examples, we construct tilings which produce ST-compatible versions of the Del Pezzo quivers and the quivers first described by Le for Fock-Goncharov coordinates for Lie groups of type $B$. We show that polygonal cluster algebras have natural evaluations in Clifford algebras, which we use to produce noncommutative generalizations of the Somos sequences and to parameterize the $\Theta$-positive semigroup of $\mathrm{Spin}(2,n)$. We indicate how this will be done for the semigroup in $\mathrm{Spin}(p,q)$ and how one will give coordinates for general $\Theta$-positive representations into $\mathrm{Spin}(p,q)$., Comment: 64 pages, 46 figures. Comments Welcome!

Current frontier video diffusion models have demonstrated remarkable results at generating high-quality videos. However, they can only generate short video clips, normally around 10 seconds or 240 frames, due to computation limitations during training. In this work, we show that existing models can be naturally extended to autoregressive video diffusion models without changing the architectures. Our key idea is to assign the latent frames with progressively increasing noise levels rather than a single noise level, which allows for fine-grained condition among the latents and large overlaps between the attention windows. Such progressive video denoising allows our models to autoregressively generate video frames without quality degradation or abrupt scene changes. We present state-of-the-art results on long video generation at 1 minute (1440 frames at 24 FPS). Videos from this paper are available at https://desaixie.github.io/pa-vdm/., Comment: 15 pages, 5 figures. Our video results and code are available at https://desaixie.github.io/pa-vdm/

Low trust remains a significant barrier to Autonomous Vehicle (AV) adoption. To design trustworthy AVs, we need to better understand the individual traits, attitudes, and experiences that impact people's trust judgements. We use machine learning to understand the most important factors that contribute to young adult trust based on a comprehensive set of personal factors gathered via survey (n = 1457). Factors ranged from psychosocial and cognitive attributes to driving style, experiences, and perceived AV risks and benefits. Using the explainable AI technique SHAP, we found that perceptions of AV risks and benefits, attitudes toward feasibility and usability, institutional trust, prior experience, and a person's mental model are the most important predictors. Surprisingly, psychosocial and many technology- and driving-specific factors were not strong predictors. Results highlight the importance of individual differences for designing trustworthy AVs for diverse groups and lead to key implications for future design and research., Comment: 24 pages (including references and appendix), Accepted to CHI Conference on Human Factors in Computing Systems (CHI 2025)

Explanations for autonomous vehicle (AV) decisions may build trust, however, explanations can contain errors. In a simulated driving study (n = 232), we tested how AV explanation errors, driving context characteristics (perceived harm and driving difficulty), and personal traits (prior trust and expertise) affected a passenger's comfort in relying on an AV, preference for control, confidence in the AV's ability, and explanation satisfaction. Errors negatively affected all outcomes. Surprisingly, despite identical driving, explanation errors reduced ratings of the AV's driving ability. Severity and potential harm amplified the negative impact of errors. Contextual harm and driving difficulty directly impacted outcome ratings and influenced the relationship between errors and outcomes. Prior trust and expertise were positively associated with outcome ratings. Results emphasize the need for accurate, contextually adaptive, and personalized AV explanations to foster trust, reliance, satisfaction, and confidence. We conclude with design, research, and deployment recommendations for trustworthy AV explanation systems., Comment: 17 pages, Accepted to CHI Conference on Human Factors in Computing Systems (CHI 2025)

Presented here is a transcription of the lecture notes from Professor Allan N. Kaufman's graduate statistical mechanics course at Berkeley from the 1972-1973 academic year. Part 1 addresses equilibrium statistical mechanics with topics: fundamentals, classical fluids and other systems, chemical equilibrium, and long-range interactions. Part 2 addresses non-equilibrium statistical mechanics with topics: fundamentals, Brownian motion, Liouville and Klimontovich equations, Landau equation, Markov processes and Fokker-Planck equation, linear response and transport theory, and an introduction to non-equilibrium quantum statistical mechanics., Comment: 165 pages, lecture notes, ten figures

We report strongly thermochromic YSZ/V0.855W0.018Sr0.127O2/SiO2 coatings, where YSZ is Y stabilized ZrO2, prepared using a scalable deposition technique on standard glass at a low substrate temperature of 320 {\deg}C and without any substrate bias voltage. The coatings exhibit a transition temperature of 22 {\deg}C with an integral luminous transmittance of 63.7% (low-temperature state) and 60.7% (high-temperature state), and a modulation of the solar energy transmittance of 11.2%. Such a combination of properties, together with the low deposition temperature, fulfill the requirements for large-scale implementation on building glass and have not been reported yet. Reactive high-power impulse magnetron sputtering with a pulsed O2 flow feedback control allows us to prepare crystalline W and Sr co-doped VO2 of the correct stoichiometry. The W doping of VO2 decreases the transition temperature, while the Sr doping of VO2 increases the luminous transmittance significantly. A coating design utilizing a second-order interference in two antireflection layers is used to maximize both the integral luminous transmittance and the modulation of the solar energy transmittance. A compact crystalline structure of the bottom YSZ antireflection layer further improves the VO2 crystallinity, while the top SiO2 antireflection layer provides also the mechanical and environmental protection for the V0.855W0.018Sr0.127O2 layer.

In this study, the authors highlight the differences in classroom-, teacher-, and school-level factors in 8th and 9th grade algebra experiences along socioeconomic and racial/ethnic lines using nationally representative survey data from the American Mathematics Educator Study. Several takeaways emerge from this analysis. The analysis shows that highest-poverty schools (i.e., schools in the top poverty quartile) are significantly less likely to offer algebra in 8th grade unconditionally (i.e., without needing to meet certain conditions) for all students or to offer algebra at all compared to lowest-poverty schools (i.e., schools in the lowest poverty quartile). Second, there are significant differences in which factors (e.g., parent requests, teacher referrals) are considered when placing students in advanced math courses in 8th and 9th grade that may affect the access of students from disadvantaged backgrounds to these courses or to more advanced pathways. Third, there are significant differences in 8th and 9th grade math teacher qualifications and classroom activities in math courses, with teachers in highest-poverty schools being significantly more likely to have received alternative credentials, less likely to have completed student-teaching during their preparation program, and less likely to have completed their state's licensure requirements for math. 8th and 9th grade math teachers in highest-poverty schools were also more likely to report that they spend more than half of their instruction time addressing math topics below grade level or addressing disciplinary issues. Mostly similar, albeit weaker, patterns emerge when discrepancies are examined along school racial/ethnic composition. Offering 8th grade algebra in high-poverty school settings (or making it available to more or all students) could help close socioeconomic gaps in algebra enrollment in 8th grade and grant more equitable access to advanced math coursework in the long-run. That said, focusing on the provision of 8th grade algebra alone will likely not remedy the opportunity gaps in access to (and completion of) advanced math courses in high school since the findings suggest that highest-poverty high schools are also significantly less likely to offer college credit-bearing math courses. Further, the findings suggest that increasing the provision of algebra in 8th grade may present three challenges: (1) staffing these courses with qualified teachers; (2) providing strong supports for students who struggle with algebra; and, relatedly, (3) making algebra placement decisions that minimize failure and maximize success for the greatest number of students. Taken together, these findings demonstrate systemic inequities across racial/ethnic and socioeconomic lines in terms of access to, and experiences in, 8th and 9th grade math courses.

Formal verification is a promising method for producing reliable software, but the difficulty of manually writing verification proofs severely limits its utility in practice. Recent methods have automated some proof synthesis by guiding a search through the proof space using a theorem prover. Unfortunately, the theorem prover provides only the crudest estimate of progress, resulting in effectively undirected search. To address this problem, we create QEDCartographer, an automated proof-synthesis tool that combines supervised and reinforcement learning to more effectively explore the proof space. QEDCartographer incorporates the proofs' branching structure, enabling reward-free search and overcoming the sparse reward problem inherent to formal verification. We evaluate QEDCartographer using the CoqGym benchmark of 68.5K theorems from 124 open-source Coq projects. QEDCartographer fully automatically proves 21.4% of the test-set theorems. Previous search-based proof-synthesis tools Tok, Tac, ASTactic, Passport, and Proverbot9001, which rely only on supervised learning, prove 9.6%, 9.8%, 10.9%, 12.5%, and 19.8%, respectively. Diva, which combines 62 tools, proves 19.2%. Comparing to the most effective prior tool, Proverbot9001, QEDCartographer produces 34% shorter proofs 29% faster, on average over the theorems both tools prove. Together, QEDCartographer and non-learning-based CoqHammer prove 30.3% of the theorems, while CoqHammer alone proves 26.6%. Our work demonstrates that reinforcement learning is a fruitful research direction for improving proof-synthesis tools' search mechanisms., Comment: Authors could not agree on final revision. Please see author websites for individual versions of paper

Contributors to open source software packages often describe feeling discouraged by the lack of positive feedback from users. This paper describes a technology probe, Hug Reports, that provides users a communication affordance within their code editors, through which users can convey appreciation to contributors of packages they use. In our field study, 18 users interacted with the probe for 3 weeks, resulting in messages of appreciation to 550 contributors, 26 of whom participated in subsequent research. Our findings show how locating a communication affordance within the code editor, and allowing users to express appreciation in terms of the abstractions they are exposed to (packages, modules, functions), can support exchanges of appreciation that are meaningful to users and contributors. Findings also revealed the moments in which users expressed appreciation, the two meanings that appreciation took on -- as a measure of utility and as an act of expressive communication -- and how contributors' reactions to appreciation were influenced by their perceived level of contribution. Based on these findings, we discuss opportunities and challenges for designing appreciation systems for open source in particular, and peer production communities more generally., Comment: To appear at CSCW 2025

We introduce NNsight and NDIF, technologies that work in tandem to enable scientific study of very large neural networks. NNsight is an open-source system that extends PyTorch to introduce deferred remote execution. NDIF is a scalable inference service that executes NNsight requests, allowing users to share GPU resources and pretrained models. These technologies are enabled by the intervention graph, an architecture developed to decouple experiment design from model runtime. Together, this framework provides transparent and efficient access to the internals of deep neural networks such as very large language models (LLMs) without imposing the cost or complexity of hosting customized models individually. We conduct a quantitative survey of the machine learning literature that reveals a growing gap in the study of the internals of large-scale AI. We demonstrate the design and use of our framework to address this gap by enabling a range of research methods on huge models. Finally, we conduct benchmarks to compare performance with previous approaches. Code documentation, and materials are available at https://nnsight.net/., Comment: Code at https://nnsight.net

Social media platforms enhance the propagation of online misinformation by providing large user bases with a quick means to share content. One way to disrupt the rapid dissemination of misinformation at scale is through warning tags, which label content as potentially false or misleading. Past warning tag mitigation studies yield mixed results for diverse audiences, however. We hypothesize that personalizing warning tags to the individual characteristics of their diverse users may enhance mitigation effectiveness. To reach the goal of personalization, we need to understand how people differ and how those differences predict a person's attitudes and self-described behaviors toward tags and tagged content. In this study, we leverage Amazon Mechanical Turk (n = 132) and undergraduate students (n = 112) to provide this foundational understanding. Specifically, we find attitudes towards warning tags and self-described behaviors are positively influenced by factors such as Personality Openness and Agreeableness, Need for Cognitive Closure (NFCC), Cognitive Reflection Test (CRT) score, and Trust in Medical Scientists. Conversely, Trust in Religious Leaders, Conscientiousness, and political conservatism were negatively correlated with these attitudes and behaviors. We synthesize our results into design insights and a future research agenda for more effective and personalized misinformation warning tags and misinformation mitigation strategies more generally., Comment: 22 pages, 6 figures

Data augmentation (DA) methods tailored to specific domains generate synthetic samples by applying transformations that are appropriate for the characteristics of the underlying data domain, such as rotations on images and time warping on time series data. In contrast, domain-independent approaches, e.g. mixup, are applicable to various data modalities, and as such they are general and versatile. While regularizing classification tasks via DA is a well-explored research topic, the effect of DA on regression problems received less attention. To bridge this gap, we study the problem of domain-independent augmentation for regression, and we introduce FOMA: a new data-driven domain-independent data augmentation method. Essentially, our approach samples new examples from the tangent planes of the train distribution. Augmenting data in this way aligns with the network tendency towards capturing the dominant features of its input signals. We evaluate FOMA on in-distribution generalization and out-of-distribution robustness benchmarks, and we show that it improves the generalization of several neural architectures. We also find that strong baselines based on mixup are less effective in comparison to our approach. Our code is publicly available athttps://github.com/azencot-group/FOMA., Comment: Accepted to ICML 2024

We present LRM-Zero, a Large Reconstruction Model (LRM) trained entirely on synthesized 3D data, achieving high-quality sparse-view 3D reconstruction. The core of LRM-Zero is our procedural 3D dataset, Zeroverse, which is automatically synthesized from simple primitive shapes with random texturing and augmentations (e.g., height fields, boolean differences, and wireframes). Unlike previous 3D datasets (e.g., Objaverse) which are often captured or crafted by humans to approximate real 3D data, Zeroverse completely ignores realistic global semantics but is rich in complex geometric and texture details that are locally similar to or even more intricate than real objects. We demonstrate that our LRM-Zero, trained with our fully synthesized Zeroverse, can achieve high visual quality in the reconstruction of real-world objects, competitive with models trained on Objaverse. We also analyze several critical design choices of Zeroverse that contribute to LRM-Zero's capability and training stability. Our work demonstrates that 3D reconstruction, one of the core tasks in 3D vision, can potentially be addressed without the semantics of real-world objects. The Zeroverse's procedural synthesis code and interactive visualization are available at: https://desaixie.github.io/lrm-zero/., Comment: 23 pages, 8 figures. Our code and interactive visualization are available at: https://desaixie.github.io/lrm-zero/. v2: NeurIPS 2024 Camera Ready version

In this paper, using Monte Carlo simulations we show that the Blume-Capel model gives rise to the social depolarization. This model borrowed from statistical physics uses the continuous Ising spin varying from -1 to 1 passing by zero to express the political stance of an individual going from ultra-left (-1) to ultra-right (+1). The particularity of the Blume-Capel model is the existence of a $D$-term which favors the state of spin zero which is a neutral stance. We consider the political system of the USA where voters affiliate with two political groups: Democrats or Republicans, or are independent. Each group is composed of a large number of interacting members of the same stance. We represent the general political ambiance (or degree of social turmoil) with a temperature $T$ similar to thermal agitation in statistical physics. When three groups interact with each other, their stances can get closer or further from each other, depending on the nature of their inter-group interactions. We study the dynamics of such variations as functions of the value of the $D$-term of each group. We show that the polarization decreases with increasing $D$. We outline the important role of $T$ in these dynamics. These MC results are in excellent agreement with the mean-field treatment of the same model., Comment: 11 pages, 9 figures