Search

We consider a general model for high-dimensional empirical risk minimization whereby the data $\mathbf{x}_i$ are $d$-dimensional isotropic Gaussian vectors, the model is parametrized by $\mathbf{\Theta}\in\mathbb{R}^{d\times k}$, and the loss depends on the data via the projection $\mathbf{\Theta}^\mathsf{T}\mathbf{x}_i$. This setting covers as special cases classical statistics methods (e.g. multinomial regression and other generalized linear models), but also two-layer fully connected neural networks with $k$ hidden neurons. We use the Kac-Rice formula from Gaussian process theory to derive a bound on the expected number of local minima of this empirical risk, under the proportional asymptotics in which $n,d\to\infty$, with $n\asymp d$. Via Markov's inequality, this bound allows to determine the positions of these minimizers (with exponential deviation bounds) and hence derive sharp asymptotics on the estimation and prediction error. In this paper, we apply our characterization to convex losses, where high-dimensional asymptotics were not (in general) rigorously established for $k\ge 2$. We show that our approach is tight and allows to prove previously conjectured results. In addition, we characterize the spectrum of the Hessian at the minimizer. A companion paper applies our general result to non-convex examples., Comment: 95 pages, 5 figures

Tokenization is fundamental to Natural Language Processing (NLP), directly impacting model efficiency and linguistic fidelity. While Byte Pair Encoding (BPE) is widely used in Large Language Models (LLMs), it often disregards morpheme boundaries, leading to suboptimal segmentation, particularly in morphologically rich languages. We introduce MorphBPE, a morphology-aware extension of BPE that integrates linguistic structure into subword tokenization while preserving statistical efficiency. Additionally, we propose two morphology-based evaluation metrics: (i) Morphological Consistency F1-Score, which quantifies the consistency between morpheme sharing and token sharing, contributing to LLM training convergence, and (ii) Morphological Edit Distance, which measures alignment between morphemes and tokens concerning interpretability. Experiments on English, Russian, Hungarian, and Arabic across 300M and 1B parameter LLMs demonstrate that MorphBPE consistently reduces cross-entropy loss, accelerates convergence, and improves morphological alignment scores. Fully compatible with existing LLM pipelines, MorphBPE requires minimal modifications for integration. The MorphBPE codebase and tokenizer playground will be available at: https://github.com/llm-lab-org/MorphBPE and https://tokenizer.llm-lab.org

Large language models (LLMs) have demonstrated remarkable proficiency in generating detailed and coherent explanations of complex concepts. However, the extent to which these models truly comprehend the concepts they articulate remains unclear. To assess the level of comprehension of a model relative to the content it generates, we implemented a self-evaluation pipeline where models: (i) given a topic generate an excerpt with information about the topic, (ii) given an excerpt generate question-answer pairs, and finally (iii) given a question generate an answer. We refer to this self-evaluation approach as Explain-Query-Test (EQT). Interestingly, the accuracy on generated questions resulting from running the EQT pipeline correlates strongly with the model performance as verified by typical benchmarks such as MMLU-Pro. In other words, EQT's performance is predictive of MMLU-Pro's, and EQT can be used to rank models without the need for any external source of evaluation data other than lists of topics of interest. Moreover, our results reveal a disparity between the models' ability to produce detailed explanations and their performance on questions related to those explanations. This gap highlights fundamental limitations in the internal knowledge representation and reasoning abilities of current LLMs. We release the code at https://github.com/asgsaeid/EQT.

The SemEval-2024 Task 3 presents two subtasks focusing on emotion-cause pair extraction within conversational contexts. Subtask 1 revolves around the extraction of textual emotion-cause pairs, where causes are defined and annotated as textual spans within the conversation. Conversely, Subtask 2 extends the analysis to encompass multimodal cues, including language, audio, and vision, acknowledging instances where causes may not be exclusively represented in the textual data. Our proposed model for emotion-cause analysis is meticulously structured into three core segments: (i) embedding extraction, (ii) cause-pair extraction & emotion classification, and (iii) cause extraction using QA after finding pairs. Leveraging state-of-the-art techniques and fine-tuning on task-specific datasets, our model effectively unravels the intricate web of conversational dynamics and extracts subtle cues signifying causality in emotional expressions. Our team, AIMA, demonstrated strong performance in the SemEval-2024 Task 3 competition. We ranked as the 10th in subtask 1 and the 6th in subtask 2 out of 23 teams., Comment: Proceedings of the 18th International Workshop on Semantic Evaluation (SemEval-2024)

In this study, we introduce a solution to the SemEval 2024 Task 10 on subtask 1, dedicated to Emotion Recognition in Conversation (ERC) in code-mixed Hindi-English conversations. ERC in code-mixed conversations presents unique challenges, as existing models are typically trained on monolingual datasets and may not perform well on code-mixed data. To address this, we propose a series of models that incorporate both the previous and future context of the current utterance, as well as the sequential information of the conversation. To facilitate the processing of code-mixed data, we developed a Hinglish-to-English translation pipeline to translate the code-mixed conversations into English. We designed four different base models, each utilizing powerful pre-trained encoders to extract features from the input but with varying architectures. By ensembling all of these models, we developed a final model that outperforms all other baselines., Comment: Proceedings of the 18th International Workshop on Semantic Evaluation (SemEval-2024)

We present Fanar, a platform for Arabic-centric multimodal generative AI systems, that supports language, speech and image generation tasks. At the heart of Fanar are Fanar Star and Fanar Prime, two highly capable Arabic Large Language Models (LLMs) that are best in the class on well established benchmarks for similar sized models. Fanar Star is a 7B (billion) parameter model that was trained from scratch on nearly 1 trillion clean and deduplicated Arabic, English and Code tokens. Fanar Prime is a 9B parameter model continually trained on the Gemma-2 9B base model on the same 1 trillion token set. Both models are concurrently deployed and designed to address different types of prompts transparently routed through a custom-built orchestrator. The Fanar platform provides many other capabilities including a customized Islamic Retrieval Augmented Generation (RAG) system for handling religious prompts, a Recency RAG for summarizing information about current or recent events that have occurred after the pre-training data cut-off date. The platform provides additional cognitive capabilities including in-house bilingual speech recognition that supports multiple Arabic dialects, voice and image generation that is fine-tuned to better reflect regional characteristics. Finally, Fanar provides an attribution service that can be used to verify the authenticity of fact based generated content. The design, development, and implementation of Fanar was entirely undertaken at Hamad Bin Khalifa University's Qatar Computing Research Institute (QCRI) and was sponsored by Qatar's Ministry of Communications and Information Technology to enable sovereign AI technology development.

Both intrinsic and extrinsic orbital Hall effects (OHE) in bilayer transition metal dichalcogenides (TMDs) are investigated in the presence of short-range disorder using quantum kinetic theory. Bilayer TMDs provide an ideal platform to study the effects of inversion symmetry breaking on transport properties due to their unique structural and electronic characteristics. While bilayer TMDs are naturally inversion symmetric, applying a finite gate voltage to create a bias between the layers effectively breaks this symmetry. Our findings reveal that slightly away from the band edges, the extrinsic OHE becomes the dominant contribution in both inversion-symmetric and asymmetric cases, with its prominence increasing significantly as the Fermi energy rises. Furthermore, we demonstrate that breaking inversion symmetry greatly enhances the extrinsic OHE. This enhancement arises from the fundamentally distinct behavior of orbital angular momentum (OAM) in centrosymmetric systems, where intraband components vanish due to symmetry constraints. As a result, in centrosymmetric systems, only the off-diagonal components of the density matrix contribute to the extrinsic OHE. In contrast, in noncentrosymmetric systems, both diagonal and off-diagonal components play a role. Our study suggests that in experimentally relevant, highly doped systems, the OHE is predominantly extrinsic in nature, regardless of whether the system is centrosymmetric or noncentrosymmetric. Importantly, we infer that even a weakly breaking of inversion symmetry can lead to a dramatic enhancement of the OHE, a finding with significant implications for experimental investigations., Comment: 12 pages, 7 figures

Handling latent variables in Structural Equation Models (SEMs) in a case where both the latent variables and their corresponding indicators in the measurement error part of the model are random curves presents significant challenges, especially with sparse data. In this paper, we develop a novel family of Functional Structural Equation Models (FSEMs) that incorporate latent variables modeled as Gaussian Processes (GPs). The introduced FSEMs are built upon functional regression models having response variables modeled as underlying GPs. The model flexibly adapts to cases when the random curves' realizations are observed only over a sparse subset of the domain, and the inferential framework is based on a restricted maximum likelihood approach. The advantage of this framework lies in its ability and flexibility in handling various data scenarios, including regularly and irregularly spaced points and thus missing data. To extract smooth estimates for the functional parameters, we employ a penalized likelihood approach that selects the smoothing parameters using a cross-validation method. We evaluate the performance of the proposed model using simulation studies and a real data example, which suggests that our model performs well in practice. The uncertainty associated with the estimates of the functional coefficients is also assessed by constructing confidence regions for each estimate. The goodness of fit indices that are commonly used to evaluate the fit of SEMs are developed for the FSEMs introduced in this paper. Overall, the proposed method is a promising approach for modeling functional data in SEMs with functional latent variables.

This study introduces CLASP (Contrastive Language-Speech Pretraining), a multilingual, multimodal representation tailored for audio-text information retrieval. CLASP leverages the synergy between spoken content and textual data. During training, we utilize our newly introduced speech-text dataset, which encompasses 15 diverse categories ranging from fiction to religion. CLASP's audio component integrates audio spectrograms with a pre-trained self-supervised speech model, while its language encoding counterpart employs a sentence encoder pre-trained on over 100 languages. This unified lightweight model bridges the gap between various modalities and languages, enhancing its effectiveness in handling and retrieving multilingual and multimodal data. Our evaluations across multiple languages demonstrate that CLASP establishes new benchmarks in HITS@1, MRR, and meanR metrics, outperforming traditional ASR-based retrieval approaches in specific scenarios., Comment: accepted at ECIR 2025

Intelligent surfaces have emerged as a promising solution to enhance coverage and mitigate signal fading in future wireless communication systems, thereby improving bandwidth and data rates. Passive intelligent surfaces, in particular, demonstrate significant potential for reducing both energy consumption and operational costs. However, their current functionalities are typically constrained to operation within a single polarization, frequency band, and angle of arrival, limiting their applicability in realistic scenarios. In this work, we introduce volumetric dielectric composites, referred to as metacrystals, which are designed through computer optimization to greatly extend the capabilities of intelligent surfaces. Owing to their three-dimensional geometry, which supports a large number of degrees of freedom, metacrystals can simultaneously provide complex multiplexing responses for multiple signals. Each signal can be characterized by distinct parameters, including polarization, angle of incidence, frequency, and/or orbital angular momentum. The proposed metacrystals feature binarized permittivity contrast between low-permittivity materials and air gaps, and can thus be fabricated using filament-based additive manufacturing, supporting frequencies up to 100 GHz. By being cost-effective and scalable, these structures are well-suited for industrial applications, such as integration into building walls, where they can redirect or absorb signals in both indoor and outdoor environments with high versatility at millimeter-wave frequencies and beyond.

Autonomous systems use independent decision-making with only limited human intervention to accomplish goals in complex and unpredictable environments. As the autonomy technologies that underpin them continue to advance, these systems will find their way into an increasing number of applications in an ever wider range of settings. If we are to deploy them to perform safety-critical or mission-critical roles, it is imperative that we have justified confidence in their safe and correct operation. Verification is the process by which such confidence is established. However, autonomous systems pose challenges to existing verification practices. This paper highlights viewpoints of the Roadmap Working Group of the IEEE Robotics and Automation Society Technical Committee for Verification of Autonomous Systems, identifying these grand challenges, and providing a vision for future research efforts that will be needed to address them.

The development of the sixth generation of communication networks (6G) has been gaining momentum over the past years, with a target of being introduced by 2030. Several initiatives worldwide are developing innovative solutions and setting the direction for the key features of these networks. Some common emerging themes are the tight integration of AI, the convergence of multiple access technologies and sustainable operation, aiming to meet stringent performance and societal requirements. To that end, we are introducing REASON - Realising Enabling Architectures and Solutions for Open Networks. The REASON project aims to address technical challenges in future network deployments, such as E2E service orchestration, sustainability, security and trust management, and policy management, utilising AI-native principles, considering multiple access technologies and cloud-native solutions. This paper presents REASON's architecture and the identified requirements for future networks. The architecture is meticulously designed for modularity, interoperability, scalability, simplified troubleshooting, flexibility, and enhanced security, taking into consideration current and future standardisation efforts, and the ease of implementation and training. It is structured into four horizontal layers: Physical Infrastructure, Network Service, Knowledge, and End-User Application, complemented by two vertical layers: Management and Orchestration, and E2E Security. This layered approach ensures a robust, adaptable framework to support the diverse and evolving requirements of 6G networks, fostering innovation and facilitating seamless integration of advanced technologies., Comment: Accepted for publication at IEEE Access

Academics increasingly acknowledge the predictive power of social media for a wide variety of events and, more specifically, for financial markets. Anecdotal and empirical findings show that cryptocurrencies are among the financial assets that have been affected by news and influencers' activities on Twitter. However, the extent to which Twitter crypto influencer's posts about trading signals and their effect on market prices is mostly unexplored. In this paper, we use LLMs to uncover buy and not-buy signals from influencers and news outlets' Twitter posts and use a VAR analysis with Granger Causality tests and cross-correlation analysis to understand how these trading signals are temporally correlated with the top nine major cryptocurrencies' prices. Overall, the results show a mixed pattern across cryptocurrencies and temporal periods. However, we found that for the top three cryptocurrencies with the highest presence within news and influencer posts, their aggregated LLM-detected trading signal over the preceding 24 hours granger-causes fluctuations in their market prices, exhibiting a lag of at least 6 hours. In addition, the results reveal fundamental differences in how influencers and news outlets cover cryptocurrencies.

PET imaging is an invaluable tool in clinical settings as it captures the functional activity of both healthy anatomy and cancerous lesions. Developing automatic lesion segmentation methods for PET images is crucial since manual lesion segmentation is laborious and prone to inter- and intra-observer variability. We propose PET-Disentangler, a 3D disentanglement method that uses a 3D UNet-like encoder-decoder architecture to disentangle disease and normal healthy anatomical features with losses for segmentation, reconstruction, and healthy component plausibility. A critic network is used to encourage the healthy latent features to match the distribution of healthy samples and thus encourages these features to not contain any lesion-related features. Our quantitative results show that PET-Disentangler is less prone to incorrectly declaring healthy and high tracer uptake regions as cancerous lesions, since such uptake pattern would be assigned to the disentangled healthy component., Comment: 4 pages, 2 figures, 1 table

Photometric galaxy surveys, despite their limited resolution along the line of sight, encode rich information about the large-scale structure (LSS) of the Universe thanks to the high number density and extensive depth of the data. However, the complicated selection effects in wide and deep surveys can potentially cause significant bias in the angular two-point correlation function (2PCF) measured from those surveys. In this paper, we measure the 2PCF from the newly published KiDS-Legacy sample. Given an $r$-band $5\sigma$ magnitude limit of $24.8$ and survey footprint of $1347$ deg$^2$, it achieves an excellent combination of sky coverage and depth for such a measurement. We find that complex selection effects, primarily induced by varying seeing, introduce over-estimation of the 2PCF by approximately an order of magnitude. To correct for such effects, we apply a machine learning-based method to recover an organised random (OR) that presents the same selection pattern as the galaxy sample. The basic idea is to find the selection-induced clustering of galaxies using a combination of self-organising maps (SOMs) and hierarchical clustering (HC). This unsupervised machine learning method is able to recover complicated selection effects without specifying their functional forms. We validate this SOM+HC method on mock deep galaxy samples with realistic systematics and selections derived from the KiDS-Legacy catalogue. Using mock data, we demonstrate that the OR delivers unbiased 2PCF cosmological parameter constraints, removing the $27\sigma$ offset in the galaxy bias parameter that is recovered when adopting uniform randoms. Blinded measurements on the real KiDS-Legacy data show that the corrected 2PCF is robust to the SOM+HC configuration near the optimal set-up suggested by the mock tests., Comment: 29 pages, 27 figures, 4 tables; Accepted for publication on Astronomy & Astrophysics; The code used for this work is published on https://github.com/yanzastro/tiaogeng

Segmenting an object in a video presents significant challenges. Each pixel must be accurately labelled, and these labels must remain consistent across frames. The difficulty increases when the segmentation is with arbitrary granularity, meaning the number of segments can vary arbitrarily, and masks are defined based on only one or a few sample images. In this paper, we address this issue by employing a pre-trained text to image diffusion model supplemented with an additional tracking mechanism. We demonstrate that our approach can effectively manage various segmentation scenarios and outperforms state-of-the-art alternatives., Comment: Technical report. Project page is at \url{https://segment-me-in-time.github.io/}

In this paper, a new heterostructure based on the hybridization of graphene-LiF layers with a nonlinear material is introduced and studied. The numerical results are depicted and discussed in detail. A high value of FOM (FOM=24.5) at the frequency of 9.22 THz is reported for the chemical potential of 0.2 ev. Our results show that the propagation features of the proposed structure can be varied by the graphene parameters and the nonlinearity inside and outside the phononic band. The Hybridization of graphene with a nonlinear medium and a polar dielectric like LiF can support high levels of confinement with low optical loss, which makes this platform a unique candidate for THz applications., Comment: 15 pages, 9 figures

This paper studies the problem of federated training of a deep neural network (DNN) over a random access channel (RACH) such as in computer networks, wireless networks, and cellular systems. More precisely, a set of remote users participate in training a centralized DNN model using SGD under the coordination of a parameter server (PS). The local model updates are transmitted from the remote users to the PS over a RACH using a slotted ALOHA protocol. The PS collects the updates from the remote users, accumulates them, and sends central model updates to the users at regular time intervals. We refer to this setting as the RACH-FL setting. The RACH-FL setting crucially addresses the problem of jointly designing a (i) client selection and (ii) gradient compression strategy which addresses the communication constraints between the remote users and the PS when transmission occurs over a RACH. For the RACH-FL setting, we propose a policy, which we term the ''age-of-gradient'' (AoG) policy in which (i) gradient sparsification is performed using top-K sparsification, (ii) the error correction is performed using memory accumulation, and (iii) the slot transmission probability is obtained by comparing the current local memory magnitude minus the magnitude of the gradient update to a threshold. Intuitively, the AoG measure of ''freshness'' of the memory state is reminiscent of the concept of age-of-information (AoI) in the context of communication theory and provides a rather natural interpretation of this policy. Numerical simulations show the superior performance of the AoG policy as compared to other RACH-FL policies.

Sparse matrix-vector multiplication (SpMV) plays a vital role in various scientific and engineering fields, from scientific computing to machine learning. Traditional general-purpose processors often fall short of their peak performance with sparse data, leading to the development of domain-specific architectures to enhance SpMV. Yet, these specialized approaches, whether tailored explicitly for SpMV or adapted from matrix-matrix multiplication accelerators, still face challenges in fully utilizing hardware resources as a result of sparsity. To tackle this problem, we introduce GUST, a hardware/software co-design, the key insight of which lies in separating multipliers and adders in the hardware, thereby enabling resource sharing across multiple rows and columns, leading to efficient hardware utilization and ameliorating negative performance impacts from sparsity. Resource sharing, however, can lead to collisions, a problem we address through a specially devised edge-coloring scheduling algorithm. Our comparisons with various prior domain specific architectures using real-world datasets shows the effectiveness of GUST, with an average hardware utilization of $33.67\%$., Comment: To Be Published In ASPLOS 2025

We present TuringQ, the first benchmark designed to evaluate the reasoning capabilities of large language models (LLMs) in the theory of computation. TuringQ consists of 4,006 undergraduate and graduate-level question-answer pairs, categorized into four difficulty levels and covering seven core theoretical areas. We evaluate several open-source LLMs, as well as GPT-4, using Chain of Thought prompting and expert human assessment. Additionally, we propose an automated LLM-based evaluation system that demonstrates competitive accuracy when compared to human evaluation. Fine-tuning a Llama3-8B model on TuringQ shows measurable improvements in reasoning ability and out-of-domain tasks such as algebra. TuringQ serves as both a benchmark and a resource for enhancing LLM performance in complex computational reasoning tasks. Our analysis offers insights into LLM capabilities and advances in AI comprehension of theoretical computer science., Comment: Accepted to EMNLP Findings 2024

We introduce OneCovariance, an open-source software designed to accurately compute covariance matrices for an arbitrary set of two-point summary statistics across a variety of large-scale structure tracers. Utilising the halo model, we estimate the statistical properties of matter and biased tracer fields, incorporating all Gaussian, non-Gaussian, and super-sample covariance terms. The flexible configuration permits user-specific parameters, such as the complexity of survey geometry, the halo occupation distribution employed to define each galaxy sample, or the form of the real-space and/or Fourier space statistics to be analysed. We illustrate the capabilities of OneCovariance within the context of a cosmic shear analysis of the final data release of the Kilo-Degree Survey (KiDS-Legacy). Upon comparing our estimated covariance with measurements from mock data and calculations from independent software, we ascertain that OneCovariance achieves accuracy at the per cent level. When assessing the impact of ignoring complex survey geometry in the cosmic shear covariance computation, we discover misestimations at approximately the $10\%$ level for cosmic variance terms. Nonetheless, these discrepancies do not significantly affect the KiDS-Legacy recovery of cosmological parameters. We derive the cross-covariance between real-space correlation functions, bandpowers, and COSEBIs, facilitating future consistency tests among these three cosmic shear statistics. Additionally, we calculate the covariance matrix of photometric-spectroscopic galaxy clustering measurements, validating Jackknife covariance estimates for calibrating KiDS-Legacy redshift distributions. The OneCovariance can be found on github Hub together with a comprehensive documentation and examples., Comment: 37 pages, 11 figures, to be submitted to A&A, code available at https://github.com/rreischke/OneCovariance

We present a novel approach to calibrating linguistic expressions of certainty, e.g., "Maybe" and "Likely". Unlike prior work that assigns a single score to each certainty phrase, we model uncertainty as distributions over the simplex to capture their semantics more accurately. To accommodate this new representation of certainty, we generalize existing measures of miscalibration and introduce a novel post-hoc calibration method. Leveraging these tools, we analyze the calibration of both humans (e.g., radiologists) and computational models (e.g., language models) and provide interpretable suggestions to improve their calibration.

Existing 3D facial emotion modeling have been constrained by limited emotion classes and insufficient datasets. This paper introduces "Emo3D", an extensive "Text-Image-Expression dataset" spanning a wide spectrum of human emotions, each paired with images and 3D blendshapes. Leveraging Large Language Models (LLMs), we generate a diverse array of textual descriptions, facilitating the capture of a broad spectrum of emotional expressions. Using this unique dataset, we conduct a comprehensive evaluation of language-based models' fine-tuning and vision-language models like Contranstive Language Image Pretraining (CLIP) for 3D facial expression synthesis. We also introduce a new evaluation metric for this task to more directly measure the conveyed emotion. Our new evaluation metric, Emo3D, demonstrates its superiority over Mean Squared Error (MSE) metrics in assessing visual-text alignment and semantic richness in 3D facial expressions associated with human emotions. "Emo3D" has great applications in animation design, virtual reality, and emotional human-computer interaction., Comment: 11 pages, 10 figures

We explore the enhanced self-calibration of photometric galaxy redshift distributions, $n(z)$, through the combination of up to six two-point functions. Our $\rm 3\times2pt$ configuration is comprised of photometric shear, spectroscopic galaxy clustering, and spectroscopic-photometric galaxy-galaxy lensing (GGL). We further include spectroscopic-photometric cross-clustering; photometric GGL; and photometric auto-clustering, using the photometric shear sample as density tracer. We perform simulated likelihood forecasts of the cosmological and nuisance parameter constraints for Stage-III- and Stage-IV-like surveys. For the Stage-III-like case, we employ realistic but perturbed redshift distributions, and distinguish between "coherent" shifting in one direction, versus more internal scattering and full-shape errors. For perfectly known $n(z)$, a $\rm 6\times2pt$ analysis gains $\sim40\%$ in Figure of Merit (FoM) in the $S_8\equiv\sigma_8\sqrt{\Omega_{\rm m}/0.3}$ and $\Omega_{\rm m}$ plane relative to the $\rm 3\times2pt$ analysis. If untreated, coherent and incoherent redshift errors lead to inaccurate inferences of $S_8$ and $\Omega_{\rm m}$, respectively. Employing bin-wise scalar shifts $\delta{z}_i$ in the tomographic mean redshifts reduces cosmological parameter biases, with a $\rm 6x2pt$ analysis constraining the shift parameters with $2-4$ times the precision of a photometric $\rm 3^{ph}\times2pt$ analysis. For the Stage-IV-like survey, a $\rm 6\times2pt$ analysis doubles the FoM($\sigma_8{-}\Omega_{\rm m}$) compared to any $\rm 3\times2pt$ or $\rm 3^{ph}\times2pt$ analysis, and is only $8\%$ less constraining than if the $n(z)$ were perfectly known. A Gaussian mixture model for the $n(z)$ reduces mean-redshift errors and preserves the $n(z)$ shape. It also yields the most accurate and precise cosmological constraints for any $N\rm\times2pt$ configuration given $n(z)$ biases., Comment: 38 pages, 20 figures, to be submitted to A&A

We construct an integral representation for the global Rankin-Selberg (partial) $L$-function $L(s, \pi \times \tau)$ where $\pi$ is an irreducible globally generic cuspidal automorphic representation of a general spin group (over an arbitrary number field) and $\tau$ is one of a general linear group, generalizing the works of Gelbart, Piatetski-Shapiro, Rallis, Ginzburg, Soudry and Kaplan among others. We consider all ranks and both even and odd general spin groups including the quasi-split forms. The resulting facts about the location of poles of $L(s, \pi \times \tau)$ have, in particular, important consequences in describing the image of the Langlands funtorial transfer from the general spin groups to general linear groups.

We study the properties of luminous red galaxies (LRGs) selected from the fourth data release of the Kilo Degree Survey (KiDS-1000) via galaxy-galaxy lensing of the background galaxies from KiDS-1000. We used a halo model formalism to interpret our measurements and obtain estimates of the halo masses as well as the satellite fractions of the LRGs, resulting in halo masses of $2.7 \times 10^{12} h^{-1} {\rm M}_{\odot}

We present a comprehensive formalism for calculating the linear and nonlinear optical response of time-periodic (Floquet) quantum systems. Our approach, based on density matrix evolution in the Floquet basis, employs the length gauge and incorporates both intraband and interband contributions of the position operator. This formalism enables the interpretation of optical responses in terms of photon-assisted transitions and reveals a unique, divergent AC response to DC fields in Floquet systems, analogous to a Drude peak at finite frequency. Importantly, our method generalizes to optical tensor conductivity calculations at arbitrary perturbation orders, providing a powerful tool for analyzing driven quantum systems. Additionally, this approach captures various DC photocurrents, including shift current, injection current, gyration current, Berry dipole contributions, and intrinsic Fermi surface effects in certain limits., Comment: 5 pages, 1 figure+Supplement (10 pages, 1 figure)

The proliferation of complex online media has accelerated the process of ideology formation, influenced by stakeholders through advertising channels. The media channels, which vary in cost and effectiveness, present a dilemma in prioritizing optimal fund allocation. There are technical challenges in describing the optimal budget allocation between channels over time, which involves defining the finite vector structure of controls on the chart. To enhance marketing productivity, it's crucial to determine how to distribute a budget across all channels to maximize business outcomes like revenue and ROI. Therefore, the strategy for media budget allocation is primarily an exercise focused on cost and achieving goals, by identifying a specific framework for a media program. Numerous researchers optimize the achievement and frequency of media selection models to aid superior planning decisions amid complexity and vast information availability. In this study, we present a planning model using the media mix model for advertising construction campaigns. Additionally, a decision-making strategy centered on FMEA identifies and prioritizes financial risk factors of the media system in companies. Despite some limitations, this research proposes a decision-making approach based on Z-number theory. To address the drawbacks of the RPN score, the suggested decision-making methodology integrates Z-SWARA and Z-WASPAS techniques with the FMEA method.

We derive the general Fresnel coefficients for reflection by incorporating the Fizeau drag effect in doped graphene, which arises from the unique behavior of its massless Dirac electrons. Using the standard Maxwell equations and constitutive relations, we analyze the influence of this relativistic phenomenon on the optical properties of doped graphene. Our study focuses on the angular shift of Brewster's angle in a structure where monolayer graphene is sandwiched between two static dielectric media. Our findings reveal that the presence of the Fizeau drag effect significantly enhances the Brewster angle shift, leading to substantial modifications in the optical characteristics of the graphene channel, including notable alterations in the reflectance spectrum. We demonstrate that this angular shift can be further amplified by increasing the drift velocities and charge densities of the electrons in graphene, offering a tunable mechanism for controlling optical behavior in graphene-based systems. The findings of this work have significant implications for the design and development of planar photonic devices that take advantage of the optical characteristics of graphene. This breakthrough creates new opportunities for the use of graphene in sophisticated photonic technologies, where exact control over the interactions between light and matter is essential., Comment: 7 pages, 5 figures

Complexity science, despite its broad scope and potential impact, has not kept pace with fields like artificial intelligence, biotechnology and social sciences in addressing ethical concerns. The field lacks a comprehensive ethical framework, leaving us, as a community, vulnerable to ethical challenges and dilemmas. Other areas have gone through similar experiences and created, with discussions and working groups, their guides, policies and recommendations. Therefore, here we highlight the critical absence of formal guidelines, dedicated ethical committees, and widespread discussions on ethics within the complexity science community. Drawing on insights from the disciplines mentioned earlier, we propose a roadmap to enhance ethical awareness and action. Our recommendations include (i) initiating supportive mechanisms to develop ethical guidelines specific to complex systems research, (ii) creating open-access resources, and (iii) fostering inclusive dialogues to ensure that complexity science can responsibly tackle societal challenges and achieve a more inclusive environment. By initiating this dialogue, we aim to encourage a necessary shift in how ethics is integrated into complexity research, positioning the field to address contemporary challenges more effectively., Comment: 7 pages, 2 figures, 1 Annexus, 1 table

Existing benchmarks for large language models (LLMs) increasingly struggle to differentiate between top-performing models, underscoring the need for more challenging evaluation frameworks. We introduce MMLU-Pro+, an enhanced benchmark building upon MMLU-Pro to assess shortcut learning and higher-order reasoning in LLMs. By incorporating questions with multiple correct answers across diverse domains, MMLU-Pro+ tests LLMs' ability to engage in complex reasoning and resist simplistic problem-solving strategies. Our results show that MMLU-Pro+ maintains MMLU-Pro's difficulty while providing a more rigorous test of model discrimination, particularly in multi-correct answer scenarios. We introduce novel metrics like shortcut selection ratio and correct pair identification ratio, offering deeper insights into model behavior and anchoring bias. Evaluations of six state-of-the-art LLMs reveal significant performance gaps, highlighting variations in reasoning abilities and bias susceptibility. We release the dataset and evaluation codes at \url{https://github.com/asgsaeid/mmlu-pro-plus}., Comment: Accepted to NeurIPS 2024, Safe Generative AI

Large foundation models have revolutionized the field, yet challenges remain in optimizing multi-modal models for specialized visual tasks. We propose a novel, generalizable methodology to identify preferred image distributions for black-box Vision-Language Models (VLMs) by measuring output consistency across varied input prompts. Applying this to different rendering types of 3D objects, we demonstrate its efficacy across various domains requiring precise interpretation of complex structures, with a focus on Computer-Aided Design (CAD) as an exemplar field. We further refine VLM outputs using in-context learning with human feedback, significantly enhancing explanation quality. To address the lack of benchmarks in specialized domains, we introduce CAD-VQA, a new dataset for evaluating VLMs on CAD-related visual question answering tasks. Our evaluation of state-of-the-art VLMs on CAD-VQA establishes baseline performance levels, providing a framework for advancing VLM capabilities in complex visual reasoning tasks across various fields requiring expert-level visual interpretation. We release the dataset and evaluation codes at \url{https://github.com/asgsaeid/cad_vqa}., Comment: Accepted to NeurIPS 2024, Safe Generative AI

Temporal networks are commonly used to model real-life phenomena. When these phenomena represent interactions and are captured at a fine-grained temporal resolution, they are modeled as link streams. Community detection is an essential network analysis task. Although many methods exist for static networks, and some methods have been developed for temporal networks represented as sequences of snapshots, few works can handle link streams. This article introduces the first adaptation of the well-known Modularity quality function to link streams. Unlike existing methods, it is independent of the time scale of analysis. After introducing the quality function, and its relation to existing static and dynamic definitions of Modularity, we show experimentally its relevance for dynamic community evaluation.

We propose a scheme aimed at achieving spatio-spectral control over spontaneous emission within a four-level atom-light coupling system interacting with optical vortices carrying orbital angular momentum (OAM). The atom comprises a ground level and two excited states coupled with two laser fields, forming a V subsystem where the upper states exclusively decay to a common fourth state via two channels. By investigating various initial states of the atom and considering the presence or absence of quantum interference in spontaneous emission channels, we analyze how the characteristics of the OAM-carrying vortex beam imprint onto the emission spectrum. The interplay between the optical vortex and the quantum system, including its environment modes, induces a wide variety of spatio-spectral behaviour, including two-dimensional spectral-peak narrowing, spectralpeak enhancement, spectral-peak suppression, and spontaneous emission reduction or quenching in the spatial azimuthal plane. Our findings shed light on the dynamics of atom-vortex beam light interactions and offer insights into the manipulation of emission properties at the quantum level.

We analytically study the inelastic lifetime of quasiparticles due to particle-particle interactions in a three-dimensional Fermi liquid approaching a density-wave instability. Using the G$_0$W approximation, we find that the softening of the dielectric function significantly enhances the quasiparticle decay rate near the instability. While the zero-temperature quasiparticle lifetime at the Fermi surface generally follows a $|\varepsilon_k-\varepsilon_{\rm F}|^{-\alpha}$ divergence with $\alpha=2$, we observe $\alpha=0.5$ at the instability point and $\alpha=1$ within the density-wave phase. Moreover, we demonstrate that the renormalization constant $Z$ is substantially suppressed as the instability is approached, enhancing the effective mass. We extend our analysis to ultra-cold Rydberg-dressed Fermi liquids, where the soft-core interactions promote density-wave instability, and find that our numerical G$_0$W results are in excellent agreement with our analytic predictions for quasiparticle lifetime, renormalization constant, and effective mass., Comment: 5.5 pages, 4 figures (Supplemental: 3.5 pages, 1 figure); Submitted

The transition from current Stage-III surveys such as the Kilo-Degree Survey (KiDS) to the increased area and redshift range of Stage IV surveys such as Euclid will significantly increase the precision of weak lensing analyses. However, with increasing precision, the accuracy of model assumptions needs to be evaluated. In this study, we quantify the impact of the correlated clustering of weak lensing source galaxies with the surrounding large-scale structure, known as source-lens clustering (SLC), which is commonly neglected. For this, we use simulated cosmological datasets with realistically distributed galaxies and measure shear correlation functions for both clustered and uniformly distributed source galaxies. Cosmological analyses are performed for both scenarios to quantify the impact of SLC on parameter inference for a KiDS-like and a Euclid-like setting. We find for Stage III surveys, SLC has a minor impact when accounting for nuisance parameters for intrinsic alignments and shifts of tomographic bins, as these nuisance parameters absorb the effect of SLC, thus changing their original meaning. For KiDS (Euclid), the inferred intrinsic alignment amplitude $A_{IA}$ changes from $0.11_{-0.46}^{+0.44}$ ($-0.009_{-0.080}^{+0.079}$) for data without SLC to $0.28_{-0.44}^{+0.42}$ ($0.022_{-0.082}^{+0.081}$) with SLC. However, fixed nuisance parameters lead to shifts in $S_8$ and $\Omega_{m}$, emphasizing the need for including SLC in the modelling. For Euclid, we find that $\sigma_8$, $\Omega_m$, and $w_0$ are shifted by $0.19$, $0.12$, and $0.12\, \sigma$, respectively, when including free nuisance parameters, and by $0.20$, $0.16$, and $0.32\,\sigma$ when fixing the nuisance parameters. Consequently, SLC on its own has only a small impact on the inferred parameter inference when using uninformative priors for nuisance parameters., Comment: 17 pages plus appendix, 10 figures, replaced by version accepted by Astronomy & Astrophysics. Abstract abridged on arXiv

Energy usage and GDP have been the subject of numerous studies over the past decades. It has been overlooked by previous studies that energy consumption correlates with economic growth in relation to GDP. This study uses threshold regression and Granger causality testing to identify GDP-dependent causality in ten OECD countries over the last 10 years. There is a significant correlation between economic growth and energy consumption. Energy consumption and short-term economic growth are statistically significantly correlated. There is a significant positive effect of energy consumption (EC) on GDP in the short run below the threshold of $10,936 USD since the coefficient is 3.34 and the p-value is 0.0252. There is a -0.0127-correlation coefficient and a 0.0327 p-value associated with GDP Granger-cause EC over the long run. EC and GDP are not causally related for GDP per capita above $10,936 USD. In the long run, GDP Granger causes EC with a coefficient of -0.1638 and a p-value of 0.0675. According to the study, sustainable development requires sustainable use of natural resources, technological investment, foreign direct investment, and gross fixed capital formation. Economic growth can be boosted while adhering to sustainability goals by implementing these recommendations.

This paper introduces a pioneering English-Azerbaijani (Arabic Script) parallel corpus, designed to bridge the technological gap in language learning and machine translation (MT) for under-resourced languages. Consisting of 548,000 parallel sentences and approximately 9 million words per language, this dataset is derived from diverse sources such as news articles and holy texts, aiming to enhance natural language processing (NLP) applications and language education technology. This corpus marks a significant step forward in the realm of linguistic resources, particularly for Turkic languages, which have lagged in the neural machine translation (NMT) revolution. By presenting the first comprehensive case study for the English-Azerbaijani (Arabic Script) language pair, this work underscores the transformative potential of NMT in low-resource contexts. The development and utilization of this corpus not only facilitate the advancement of machine translation systems tailored for specific linguistic needs but also promote inclusive language learning through technology. The findings demonstrate the corpus's effectiveness in training deep learning MT systems and underscore its role as an essential asset for researchers and educators aiming to foster bilingual education and multilingual communication. This research covers the way for future explorations into NMT applications for languages lacking substantial digital resources, thereby enhancing global language education frameworks. The Python package of our code is available at https://pypi.org/project/chevir-kartalol/, and we also have a website accessible at https://translate.kartalol.com/., Comment: This paper is accepted and published at NeTTT 2024 Conf

The increasing size of large language models (LLMs) challenges their usage on resource-constrained platforms. For example, memory on modern GPUs is insufficient to hold LLMs that are hundreds of Gigabytes in size. Offloading is a popular method to escape this constraint by storing weights of an LLM model to host CPU memory and SSD, then loading each weight to GPU before every use. In our case study of offloaded inference, we found that due to the low bandwidth between storage devices and GPU, the latency of transferring large model weights from its offloaded location to GPU memory becomes the critical bottleneck with actual compute taking nearly 0% of runtime. To effectively reduce the weight transfer latency, we propose a novel sparse format that compresses the unstructured sparse pattern of pruned LLM weights to non-zero values with high compression ratio and low decompression overhead. Endor achieves this by expressing the positions of non-zero elements with a bitmap. Compared to offloaded inference using the popular Huggingface Accelerate, applying Endor accelerates OPT-66B by 1.70x and Llama2-70B by 1.78x. When direct weight transfer from SSD to GPU is leveraged, Endor achieves 2.25x speedup on OPT-66B and 2.37x speedup on Llama2-70B., Comment: 14 pages, 16 figures

Computational notebooks have become the primary coding environment for data scientists. However, research on their code quality is still emerging, and the code shared is often of poor quality. Given the importance of maintenance and reusability, understanding the metrics that affect notebook code comprehensibility is crucial. Code understandability, a qualitative variable, is closely tied to user opinions. Traditional approaches to measuring it either use limited questionnaires to review a few code pieces or rely on metadata such as likes and votes in software repositories. Our approach enhances the measurement of Jupyter notebook understandability by leveraging user comments related to code understandability. As a case study, we used 542,051 Kaggle Jupyter notebooks from our previous research, named DistilKaggle. We employed a fine-tuned DistilBERT transformer to identify user comments associated with code understandability. We established a criterion called User Opinion Code Understandability (UOCU), which considers the number of relevant comments, upvotes on those comments, total notebook views, and total notebook upvotes. UOCU proved to be more effective than previous methods. Furthermore, we trained machine learning models to predict notebook code understandability based solely on their metrics. We collected 34 metrics for 132,723 final notebooks as features in our dataset, using UOCU as the label. Our predictive model, using the Random Forest classifier, achieved 89% accuracy in predicting the understandability levels of computational notebooks.