Search

This paper proposes a new hybrid algorithm, combining FA, SSO, and the N-R method to accelerate convergence towards global optima, named the Hybrid Firefly Algorithm and Sperm Swarm Optimization with Newton-Raphson (HFASSON). The performance of HFASSON is evaluated using 23 benchmark functions from the CEC 2017 suite, tested in 30, 50, and 100 dimensions. A statistical comparison is performed to assess the effectiveness of HFASSON against FA, SSO, HFASSO, and five hybrid algorithms: Water Cycle Moth Flame Optimization (WCMFO), Hybrid Particle Swarm Optimization and Genetic Algorithm (HPSOGA), Hybrid Sperm Swarm Optimization and Gravitational Search Algorithm (HSSOGSA), Grey Wolf and Cuckoo Search Algorithm (GWOCS), and Hybrid Firefly Genetic Algorithm (FAGA). Results from the Friedman rank test show the superior performance of HFASSON. Additionally, HFASSON is applied to Cognitive Radio Vehicular Ad-hoc Networks (CR-VANET), outperforming basic CR-VANET in spectrum utilization. These findings demonstrate HFASSON's efficiency in wireless network applications.

Amid the swift progress of large language models (LLMs) and their evolution into large multimodal models (LMMs), significant strides have been made in high-resource languages such as English and Chinese. While Arabic LLMs have seen notable progress, Arabic LMMs remain largely unexplored, often narrowly focusing on a few specific aspects of the language and visual understanding. To bridge this gap, we introduce AIN-the Arabic Inclusive Multimodal Model-designed to excel across diverse domains. AIN is an English-Arabic bilingual LMM designed to excel in English and Arabic, leveraging carefully constructed 3.6 million high-quality Arabic-English multimodal data samples. AIN demonstrates state-of-the-art Arabic performance, while also possessing strong English-language visual capabilities. On the recent CAMEL-Bench benchmark comprising 38 sub-domains including, multi-image understanding, complex visual perception, handwritten document understanding, video understanding, medical imaging, plant diseases, and remote sensing-based land use understanding, our AIN demonstrates strong performance with the 7B model outperforming GPT-4o by an absolute gain of 3.4% averaged over eight domains and 38 sub-domains. AIN's superior capabilities position it as a significant step toward empowering Arabic speakers with advanced multimodal generative AI tools across diverse applications., Comment: 20 pages, 16 figures, ACL

These extended lecture notes survey a novel derivation of anyonic topological order (as seen in fractional quantum Hall systems) on single magnetized M5-branes probing Seifert orbi-singularities ("geometric engineering" of anyons), which we motivate from fundamental open problems in the field of quantum computing. The rigorous construction is non-Lagrangian and non-perturbative, based on previously neglected global completion of the M5-brane's tensor field by flux-quantization consistent with its non-linear self-duality and its twisting by the bulk C-field. This exists only in little-studied non-abelian generalized cohomology theories, notably in a twisted equivariant (and "twistorial") form of unstable Cohomotopy ("Hypothesis H"). As a result, topological quantum observables form Pontrjagin homology algebras of mapping spaces from the orbi-fixed worldvolume into a classifying 2-sphere. Remarkably, results from algebraic topology imply from this the quantum observables and modular functor of abelian Chern-Simons theory, as well as braid group actions on defect anyons of the kind envisioned as hardware for topologically protected quantum gates., Comment: 41 pages: extended lecture notes, based on parts of the course "Introduction to Hypothesis H", held at 45th Srni Winter School GEOMETRY AND PHYSICS, Jan 2025

Reasoning is a fundamental capability for solving complex multi-step problems, particularly in visual contexts where sequential step-wise understanding is essential. Existing approaches lack a comprehensive framework for evaluating visual reasoning and do not emphasize step-wise problem-solving. To this end, we propose a comprehensive framework for advancing step-by-step visual reasoning in large language models (LMMs) through three key contributions. First, we introduce a visual reasoning benchmark specifically designed to evaluate multi-step reasoning tasks. The benchmark presents a diverse set of challenges with eight different categories ranging from complex visual perception to scientific reasoning with over 4k reasoning steps in total, enabling robust evaluation of LLMs' abilities to perform accurate and interpretable visual reasoning across multiple steps. Second, we propose a novel metric that assesses visual reasoning quality at the granularity of individual steps, emphasizing both correctness and logical coherence. The proposed metric offers deeper insights into reasoning performance compared to traditional end-task accuracy metrics. Third, we present a new multimodal visual reasoning model, named LlamaV-o1, trained using a multi-step curriculum learning approach, where tasks are progressively organized to facilitate incremental skill acquisition and problem-solving. The proposed LlamaV-o1 is designed for multi-step reasoning and learns step-by-step through a structured training paradigm. Extensive experiments show that our LlamaV-o1 outperforms existing open-source models and performs favorably against close-source proprietary models. Compared to the recent Llava-CoT, our LlamaV-o1 achieves an average score of 67.3 with an absolute gain of 3.8\% across six benchmarks while being 5 times faster during inference scaling. Our benchmark, model, and code are publicly available., Comment: 15 pages, 5 Figures

Large Language Models (LLMs) have become essential tools across various domains due to their impressive capabilities in understanding and generating human-like text. The ability to accurately answer multiple-choice questions (MCQs) holds significant value in education, particularly in automated tutoring systems and assessment platforms. However, adapting LLMs to handle MCQ tasks effectively remains challenging due to the hallucinations and unclear prompts. This work explores the potential of Microsoft's PHI-3\cite{Abdin2024}, a compact yet efficient LLM, for MCQ answering. Our contributions include fine-tuning the model on the TruthfulQA dataset, designing optimized prompts to enhance model performance, and evaluating using perplexity and traditional metrics like accuracy and F1 score. Results show a remarkable improvement in PHI-3.5's MCQ handling post-fine-tuning, with perplexity decreasing from 4.68 to 2.27, and accuracy rising from 62\% to 90.8\%. This research underlines the importance of efficient models in adaptive learning systems and educational assessments, paving the way for broader integration into the classroom, particularly in fields like test preparation, student feedback, and personalized learning.

This paper introduces BiMediX2, a bilingual (Arabic-English) Bio-Medical EXpert Large Multimodal Model (LMM) with a unified architecture that integrates text and visual modalities, enabling advanced image understanding and medical applications. BiMediX2 leverages the Llama3.1 architecture and integrates text and visual capabilities to facilitate seamless interactions in both English and Arabic, supporting text-based inputs and multi-turn conversations involving medical images. The model is trained on an extensive bilingual healthcare dataset consisting of 1.6M samples of diverse medical interactions for both text and image modalities, mixed in Arabic and English. We also propose the first bilingual GPT-4o based medical LMM benchmark named BiMed-MBench. BiMediX2 is benchmarked on both text-based and image-based tasks, achieving state-of-the-art performance across several medical benchmarks. It outperforms recent state-of-the-art models in medical LLM evaluation benchmarks. Our model also sets a new benchmark in multimodal medical evaluations with over 9% improvement in English and over 20% in Arabic evaluations. Additionally, it surpasses GPT-4 by around 9% in UPHILL factual accuracy evaluations and excels in various medical Visual Question Answering, Report Generation, and Report Summarization tasks. The project page including source code and the trained model, is available at https://github.com/mbzuai-oryx/BiMediX2.

Language-based agentic systems have shown great promise in recent years, transitioning from solving small-scale research problems to being deployed in challenging real-world tasks. However, optimizing these systems often requires substantial manual labor. Recent studies have demonstrated that these systems can be represented as computational graphs, enabling automatic optimization. Despite these advancements, most current efforts in Graph-based Agentic System Optimization (GASO) fail to properly assign feedback to the system's components given feedback on the system's output. To address this challenge, we formalize the concept of semantic backpropagation with semantic gradients -- a generalization that aligns several key optimization techniques, including reverse-mode automatic differentiation and the more recent TextGrad by exploiting the relationship among nodes with a common successor. This serves as a method for computing directional information about how changes to each component of an agentic system might improve the system's output. To use these gradients, we propose a method called semantic gradient descent which enables us to solve GASO effectively. Our results on both BIG-Bench Hard and GSM8K show that our approach outperforms existing state-of-the-art methods for solving GASO problems. A detailed ablation study on the LIAR dataset demonstrates the parsimonious nature of our method. A full copy of our implementation is publicly available at https://github.com/HishamAlyahya/semantic_backprop, Comment: 11 pages in main text + 2 pages of references + 15 pages of appendices, 2 figures in main text + 17 figures in appendices, 2 tables in main text + 1 table in appendices, 2 algorithms in main text; source code available at https://github.com/HishamAlyahya/semantic_backprop

In the era of foundation models, CLIP has emerged as a powerful tool for aligning text and visual modalities into a common embedding space. However, the alignment objective used to train CLIP often results in subpar visual features for fine-grained tasks. In contrast, SSL-pretrained models like DINO excel at extracting rich visual features due to their specialized training paradigm. Yet, these SSL models require an additional supervised linear probing step, which relies on fully labeled data which is often expensive and difficult to obtain at scale. In this paper, we propose a label-free prompt-tuning method that leverages the rich visual features of self-supervised learning models (DINO) and the broad textual knowledge of large language models (LLMs) to largely enhance CLIP-based image classification performance using unlabeled images. Our approach unfolds in three key steps: (1) We generate robust textual feature embeddings that more accurately represent object classes by leveraging class-specific descriptions from LLMs, enabling more effective zero-shot classification compared to CLIP's default name-specific prompts. (2) These textual embeddings are then used to produce pseudo-labels to train an alignment module that integrates the complementary strengths of LLM description-based textual embeddings and DINO's visual features. (3) Finally, we prompt-tune CLIP's vision encoder through DINO-assisted supervision using the trained alignment module. This three-step process allows us to harness the best of visual and textual foundation models, resulting in a powerful and efficient approach that surpasses state-of-the-art label-free classification methods. Notably, our framework, NoLA (No Labels Attached), achieves an average absolute gain of 3.6% over the state-of-the-art LaFter across 11 diverse image classification datasets.

We observe that there is a rigorous derivation of (abelian) anyonic quantum states, hence of "topological order", on the 1+2-dimensional fixed locus of M5-probes wrapped over a trivially Seifert-fibered 3-orbifold singularity. Similar statements have previously been conjectured by appeal to the unknown dynamics of "coincident" M5-branes, but neglecting effects of flux-quantization that, as we highlight, entail anyonic solitons already in the rigorously tractable case of single M5-brane probes. This is possible after globally completing the "self-dual" tensor field on probe M5-branes by flux quantization in the non-abelian cohomology theory called equivariant twistorial Cohomotopy, which is admissible by recent results., Comment: 15 pages

Existing Large Multimodal Models (LMMs) generally focus on only a few regions and languages. As LMMs continue to improve, it is increasingly important to ensure they understand cultural contexts, respect local sensitivities, and support low-resource languages, all while effectively integrating corresponding visual cues. In pursuit of culturally diverse global multimodal models, our proposed All Languages Matter Benchmark (ALM-bench) represents the largest and most comprehensive effort to date for evaluating LMMs across 100 languages. ALM-bench challenges existing models by testing their ability to understand and reason about culturally diverse images paired with text in various languages, including many low-resource languages traditionally underrepresented in LMM research. The benchmark offers a robust and nuanced evaluation framework featuring various question formats, including true/false, multiple choice, and open-ended questions, which are further divided into short and long-answer categories. ALM-bench design ensures a comprehensive assessment of a model's ability to handle varied levels of difficulty in visual and linguistic reasoning. To capture the rich tapestry of global cultures, ALM-bench carefully curates content from 13 distinct cultural aspects, ranging from traditions and rituals to famous personalities and celebrations. Through this, ALM-bench not only provides a rigorous testing ground for state-of-the-art open and closed-source LMMs but also highlights the importance of cultural and linguistic inclusivity, encouraging the development of models that can serve diverse global populations effectively. Our benchmark is publicly available., Comment: A Multilingual Multimodal cultural benchmark for 100 languages

Self-Admitted Technical Debt (SATD) refers to the phenomenon where developers explicitly acknowledge technical debt through comments in the source code. While considerable research has focused on detecting and addressing SATD, its true impact on software maintenance remains underexplored. The few studies that have examined this critical aspect have not provided concrete evidence linking SATD to negative effects on software maintenance. These studies, however, focused only on file- or class-level code granularity. This paper aims to empirically investigate the influence of SATD on various facets of software maintenance at the method level. We assess SATD's effects on code quality, bug susceptibility, change frequency, and the time practitioners typically take to resolve SATD. By analyzing a dataset of 774,051 methods from 49 open-source projects, we discovered that methods containing SATD are not only larger and more complex but also exhibit lower readability and a higher tendency for bugs and changes. We also found that SATD often remains unresolved for extended periods, adversely affecting code quality and maintainability. Our results provide empirical evidence highlighting the necessity of early identification, resource allocation, and proactive management of SATD to mitigate its long-term impacts on software quality and maintenance costs.

Following arguments that the (hidden) M-algebra serves as the maximal super-exceptional tangent space for 11D supergravity, we make explicit here its integration to a (super-Lie) group. This is equipped with a left-invariant extension of the ''decomposed'' M-theory 3-form, such that it constitutes the Kleinian space on which super-exceptional spacetimes are to be locally modeled as Cartan geometries. As a simple but consequential application, we highlight how to describe lattice subgroups $\mathbb{Z}^{k \leq 528}$ of the hidden M-group that allow to toroidially compactify also the ''hidden'' dimensions of a super-exceptional spacetime, akin to the familiar situation in topological T-duality. In order to deal with subtleties in these constructions, we (i) provide a computer-checked re-derivation of the ''decompose'' M-theory 3-form, and (ii) present a streamlined conception of super-Lie groups, that is both rigorous while still close to physics intuition and practice. Thereby this article highlights modernized super-Lie theory along the example of the hidden M-algebra, with an eye towards laying foundations for super-exceptional geometry. Among new observations is the dimensional reduction of the hidden M-algebra to a ''hidden IIA-algebra'' which in a companion article we explain as the exceptional extension of the T-duality doubled super-spacetime., Comment: 40 pages

Super $L_\infty$-algebras unify extended super-symmetry with rational classifying spaces for higher flux densities: The super-invariant super-fluxes which control super $p$-branes and their supergravity target super-spaces are, together with their (non-linear) Bianchi identities, neatly encoded in (non-abelian) super-$L_\infty$ cocycles. These are the rational shadows of flux-quantization laws (in ordinary cohomology, K-theory, Cohomotopy, iterated K-theory, etc). We first review, in streamlined form while filling some previous gaps, double-dimensional reduction/oxidation and 10D superspace T-duality along higher-dimensional super-tori. We do so tangent super-space wise, by viewing it as an instance of adjunctions (dualities) between super-$L_\infty$-extensions and -cyclifications, applied to the avatar super-flux densities of 10D supergravity. In particular, this yields a derivation, at the rational level, of the traditional laws of "topological T-duality" from the super-$L_\infty$ structure of type II superspace. At this level, we also discuss a higher categorical analog of T-duality involving M-branes. Then, by considering super-space T-duality along all 1+9 spacetime dimensions while retaining the 11th dimension as in F-theory, we find the M-algebra appearing as the complete brane-charge extension of the fully T-doubled/correspondence super-spacetime. On this backdrop, we recognize the "decomposed" M-theory 3-form on the "hidden M-algebra" as an M-theoretic lift of the Poincar\'e super 2-form that controls superspace T-duality as the integral kernel of the super Fourier-Mukai transform. This provides the super-space structure of an M-theory lift of the doubled/correspondence space geometry, which controls T-duality., Comment: 90 pages, some figures

The conjectured symmetries of M-theory famously involve (1.) brane-extended super-symmetry (the M-algebra) and (2.) exceptional duality-symmetry (the $\mathfrak{e}_{11}$-algebra); but little attention has been given to their inevitable combination. In this little note, we highlight (by combining results available in the literature) that the $local$ exceptional duality-symmetry (the hyperbolic involutory $\mathfrak{k}_{1,10} \subset \mathfrak{e}_{11}$) acts on the M-algebra, through the "brane-rotating symmetry" $\mathfrak{sl}_{32}$, in a way which extends the known hierarchy of finite-dimensional $n$-exceptional tangent spaces compatibly beyond the traditional bound of $n \leq 7$ all the way to $n = 11$., Comment: 13 pages; v2: published version

Student disengagement in online learning has become a critical challenge, particularly post-pandemic. This review explores deep learning techniques used to detect disengagement, emphasizing computer vision and affective computing as effective approaches. We examine recent studies focusing on facial expressions, eye movements, and posture to assess student attention, along with non-face-based indicators like mouse activity. A systematic review of 38 selected studies outlines the indicators, methods, and models employed in this field, providing insights for future research on real-time engagement monitoring in online classrooms

A comprehensive investigation of the entanglement characteristics is carried out on tripartite spin-1/2 systems, examining prototypical tripartite states, the thermal Heisenberg model, and the transverse field Ising model. The entanglement is computed using the R\'enyi relative entropy. In the traditional R\'enyi relative entropy, the generalization parameter $\alpha$ can take values only in the range $0 \leq \alpha \leq 2$ due to the requirements of joint convexity of the measure. To use the R\'enyi relative entropy over a wider range of $\alpha$, we use the sandwiched form which is jointly convex in the regime $0.5 \leq \alpha \leq \infty$. In prototypical tripartite states, we find that GHZ states are monogamous, but surprisingly so are W states. On the other hand, star states exhibit polygamy, due to the higher level of purity of the bipartite subsystems. For spin models, we study the dependence of entanglement on various parameters such as temperature, spin-spin interaction, and anisotropy, and identify regions where entanglement is the largest. The R\'enyi parameter $\alpha$ scales the amount of entanglement in the system. The entanglement measure based on the traditional and the sandwiched R\'enyi relative entropies obey the Araki-Lieb-Thirring inequality. In the Heisenberg models, namely the XYZ, XXZ, and XY models, the system is always monogamous. However, in the transverse field Ising model, the state is initially polygamous and becomes monogamous with temperature and coupling., Comment: 24 Pages

We first give a brief exposition of our recent realization of anyonic quantum states on single M5-brane probes in 11D super-gravity backgrounds, by non-perturbative quantization of the topological sector of the self-dual tensor field on the 6D worldvolume, after its proper flux-quantization. This opens the prospect of holographic models for topological qbits away from the usual but unrealistic limit of large numbers of branes. At the same time, the elementary homotopy-theoretic nature of the construction yields a slick expression of topological quantum gates in homotopically-typed programming languages, opening the prospect of topological-hardware aware quantum programming. In view of these results, we end with some more meta-physical remarks on (cohesive) homotopy (type) theory in view of emergent fundamental physics and, possibly, M-theory., Comment: 16 pages, some figures -- invited contribution to: Arsiwalla, Elshatlawy & Rickles (eds.) "Quantum Gravity and Computation"

All rings are commutative, and all modules are unital. The purpose of this paper is to investigate the characterizations of weakly pseudo primary 2-absorbing sub-module in terms of some types of modules. We provide characterizations for the class of multiplication modules with the help of some types of modules such as faithful, non-singular, Z-regular, and projective modules. Furthermore, we add some conditions to prove the residual of a weakly pseudo primary 2-absorbing sub-module is a weakly pseudo primary 2-absorbing ideal., Comment: 12 pages

Volumetric medical image segmentation is a fundamental problem in medical image analysis where the objective is to accurately classify a given 3D volumetric medical image with voxel-level precision. In this work, we propose a novel hierarchical encoder-decoder-based framework that strives to explicitly capture the local and global dependencies for volumetric 3D medical image segmentation. The proposed framework exploits local volume-based self-attention to encode the local dependencies at high resolution and introduces a novel volumetric MLP-mixer to capture the global dependencies at low-resolution feature representations, respectively. The proposed volumetric MLP-mixer learns better associations among volumetric feature representations. These explicit local and global feature representations contribute to better learning of the shape-boundary characteristics of the organs. Extensive experiments on three different datasets reveal that the proposed method achieves favorable performance compared to state-of-the-art approaches. On the challenging Synapse Multi-organ dataset, the proposed method achieves an absolute 3.82\% gain over the state-of-the-art approaches in terms of HD95 evaluation metrics {while a similar improvement pattern is exhibited in MSD Liver and Pancreas tumor datasets}. We also provide a detailed comparison between recent architectural design choices in the 2D computer vision literature by adapting them for the problem of 3D medical image segmentation. Finally, our experiments on the ZebraFish 3D cell membrane dataset having limited training data demonstrate the superior transfer learning capabilities of the proposed vMixer model on the challenging 3D cell instance segmentation task, where accurate boundary prediction plays a vital role in distinguishing individual cell instances.

We introduce Controller-Embedded Language Model Interactions (CELI), a framework that integrates control logic directly within language model (LM) prompts, facilitating complex, multi-stage task execution. CELI addresses limitations of existing prompt engineering and workflow optimization techniques by embedding control logic directly within the operational context of language models, enabling dynamic adaptation to evolving task requirements. Our framework transfers control from the traditional programming execution environment to the LMs, allowing them to autonomously manage computational workflows while maintaining seamless interaction with external systems and functions. CELI supports arbitrary function calls with variable arguments, bridging the gap between LMs' adaptive reasoning capabilities and conventional software paradigms' structured control mechanisms. To evaluate CELI's versatility and effectiveness, we conducted case studies in two distinct domains: code generation (HumanEval benchmark) and multi-stage content generation (Wikipedia-style articles). The results demonstrate notable performance improvements across a range of domains. CELI achieved a 4.9 percentage point improvement over the best reported score of the baseline GPT-4 model on the HumanEval code generation benchmark. In multi-stage content generation, 94.4% of CELI-produced Wikipedia-style articles met or exceeded first draft quality when optimally configured, with 44.4% achieving high quality. These outcomes underscore CELI's potential for optimizing AI-driven workflows across diverse computational domains., Comment: 26 pages, 2 figures

This paper addresses privacy protection in decentralized Artificial Intelligence (AI) using Confidential Computing (CC) within the Atoma Network, a decentralized AI platform designed for the Web3 domain. Decentralized AI distributes AI services among multiple entities without centralized oversight, fostering transparency and robustness. However, this structure introduces significant privacy challenges, as sensitive assets such as proprietary models and personal data may be exposed to untrusted participants. Cryptography-based privacy protection techniques such as zero-knowledge machine learning (zkML) suffers prohibitive computational overhead. To address the limitation, we propose leveraging Confidential Computing (CC). Confidential Computing leverages hardware-based Trusted Execution Environments (TEEs) to provide isolation for processing sensitive data, ensuring that both model parameters and user data remain secure, even in decentralized, potentially untrusted environments. While TEEs face a few limitations, we believe they can bridge the privacy gap in decentralized AI. We explore how we can integrate TEEs into Atoma's decentralized framework.

Significant progress has been made in advancing large multimodal conversational models (LMMs), capitalizing on vast repositories of image-text data available online. Despite this progress, these models often encounter substantial domain gaps, hindering their ability to engage in complex conversations across new domains. Recent efforts have aimed to mitigate this issue, albeit relying on domain-specific image-text data to curate instruction-tuning data. However, many domains, such as agriculture, lack such vision-language data. In this work, we propose an approach to construct instruction-tuning data that harnesses vision-only data for the agriculture domain. We utilize diverse agricultural datasets spanning multiple domains, curate class-specific information, and employ large language models (LLMs) to construct an expert-tuning set, resulting in a 70k expert-tuning dataset called AgroInstruct. Subsequently, we expert-tuned and created AgroGPT, an efficient LMM that can hold complex agriculture-related conversations and provide useful insights. We also develop AgroEvals for evaluation and compare {AgroGPT's} performance with large open and closed-source models. {AgroGPT} excels at identifying fine-grained agricultural concepts, can act as an agriculture expert, and provides helpful information for multimodal agriculture questions. The code, datasets, and models are available at https://github.com/awaisrauf/agroGPT., Comment: Accepted at WACV, 2025

Robots can greatly enhance human capabilities, yet their development presents a range of challenges. This collaborative study, conducted by a team of software engineering and robotics researchers, seeks to identify the challenges encountered by robot developers by analyzing questions posted on StackOverflow. We created a filtered dataset of 500 robotics-related questions and examined their characteristics, comparing them with randomly selected questions from the platform. Our findings indicate that the small size of the robotics community limits the visibility of these questions, resulting in fewer responses. While the number of robotics questions has been steadily increasing, they remain less popular than the average question and answer on StackOverflow. This underscores the importance of research that focuses on the challenges faced by robotics practitioners. Consequently, we conducted a thematic analysis of the 500 robotics questions to uncover common inquiry patterns. We identified 11 major themes, with questions about robot movement being the most frequent. Our analysis of yearly trends revealed that certain themes, such as Specifications, were prominent from 2009 to 2014 but have since diminished in relevance. In contrast, themes like Moving, Actuator, and Remote have consistently dominated discussions over the years. These findings suggest that challenges in robotics may vary over time. Notably, the majority of robotics questions are framed as How questions, rather than Why or What questions, revealing the lack of enough resources for the practitioners. These insights can help guide researchers and educators in developing effective and timely educational materials for robotics practitioners.

We report the synthesis of a new functional molecule, a quater-tiophene based surfactant, which can both adsorb at the water / gas interface (surface active molecule) and aggregates through pi-pi stacking interactions. We assess then the ability of this molecule to create these functionalities at interfaces. This interfacial functional aggregation, characterized here in situ for the first time, is probed thanks to Langmuir trough experiments and spectrometric ellipsometry. These results open some new routes for the design of new water based opto-electronic devices., Comment: 17 pages, 14 figures

Latent variable collaborative filtering methods have been a standard approach to modelling user-click interactions due to their simplicity and effectiveness. However, there is limited work on analyzing the mathematical properties of these methods in particular on preventing the overfitting towards the identity, and such methods typically utilize loss functions that overlook the geometry between items. In this work, we introduce a notion of generalization gap in collaborative filtering and analyze this with respect to latent collaborative filtering models. We present a geometric upper bound that gives rise to loss functions, and a way to meaningfully utilize the geometry of item-metadata to improve recommendations. We show how these losses can be minimized and gives the recipe to a new latent collaborative filtering algorithm, which we refer to as GeoCF, due to the geometric nature of our results. We then show experimentally that our proposed GeoCF algorithm can outperform other all existing methods on the Movielens20M and Netflix datasets, as well as two large-scale internal datasets. In summary, our work proposes a theoretically sound method which paves a way to better understand generalization of collaborative filtering at large., Comment: 13 pages, 1 figure, 3 tables

3D multi-object tracking plays a critical role in autonomous driving by enabling the real-time monitoring and prediction of multiple objects' movements. Traditional 3D tracking systems are typically constrained by predefined object categories, limiting their adaptability to novel, unseen objects in dynamic environments. To address this limitation, we introduce open-vocabulary 3D tracking, which extends the scope of 3D tracking to include objects beyond predefined categories. We formulate the problem of open-vocabulary 3D tracking and introduce dataset splits designed to represent various open-vocabulary scenarios. We propose a novel approach that integrates open-vocabulary capabilities into a 3D tracking framework, allowing for generalization to unseen object classes. Our method effectively reduces the performance gap between tracking known and novel objects through strategic adaptation. Experimental results demonstrate the robustness and adaptability of our method in diverse outdoor driving scenarios. To the best of our knowledge, this work is the first to address open-vocabulary 3D tracking, presenting a significant advancement for autonomous systems in real-world settings. Code, trained models, and dataset splits are available publicly., Comment: 7 pages, 4 figures, 3 tables

Large multimodal models (LMMs) have shown encouraging performance in the natural image domain using visual instruction tuning. However, these LMMs struggle to describe the content of remote sensing images for tasks such as image or region grounding, classification, etc. Recently, GeoChat make an effort to describe the contents of the RS images. Although, GeoChat achieves promising performance for various RS tasks, it struggles to describe the changes between bi-temporal RS images which is a key RS task. This necessitates the development of an LMM that can describe the changes between the bi-temporal RS images. However, there is insufficiency of datasets that can be utilized to tune LMMs. In order to achieve this, we introduce a change description instruction dataset that can be utilized to finetune an LMM and provide better change descriptions for RS images. Furthermore, we show that the LLaVA-1.5 model, with slight modifications, can be finetuned on the change description instruction dataset and achieve favorably better performance.

Inter-image association modeling is crucial for co-salient object detection. Despite satisfactory performance, previous methods still have limitations on sufficient inter-image association modeling. Because most of them focus on image feature optimization under the guidance of heuristically calculated raw inter-image associations. They directly rely on raw associations which are not reliable in complex scenarios, and their image feature optimization approach is not explicit for inter-image association modeling. To alleviate these limitations, this paper proposes a deep association learning strategy that deploys deep networks on raw associations to explicitly transform them into deep association features. Specifically, we first create hyperassociations to collect dense pixel-pair-wise raw associations and then deploys deep aggregation networks on them. We design a progressive association generation module for this purpose with additional enhancement of the hyperassociation calculation. More importantly, we propose a correspondence-induced association condensation module that introduces a pretext task, i.e. semantic correspondence estimation, to condense the hyperassociations for computational burden reduction and noise elimination. We also design an object-aware cycle consistency loss for high-quality correspondence estimations. Experimental results in three benchmark datasets demonstrate the remarkable effectiveness of our proposed method with various training settings., Comment: There is an error. In Sec 4.1, the number of images in some dataset is incorrect and needs to be revised

In previous work, we introduced Mysterious Triality, extending the Mysterious Duality of Iqbal, Neitzke, and Vafa between physics and algebraic geometry to include algebraic topology in the form of rational homotopy theory. Starting with the rational Sullivan minimal model of the 4-sphere $S^4$, capturing the dynamics of M-theory via Hypothesis H, this progresses to the dimensional reduction of M-theory on torus $T^k$, $k \ge 1$, with its dynamics described via the iterated cyclic loop space $\mathcal{L}_c^k S^4$ of the 4-sphere. From this, we also extracted data corresponding to the maximal torus/Cartan subalgebra and the Weyl group of the exceptional Lie group/algebra of type $E_k$. In this paper, we discover much richer symmetry by extending the data of the Cartan subalgebra to a maximal parabolic subalgebra $\mathfrak{p}_k^{k(k)}$ of the split real form $\mathfrak{e}_{k(k)}$ of the exceptional Lie algebra of type $E_k$ by exhibiting an action, in rational homotopy category, of $\mathfrak{p}_k^{k(k)}$ on the slightly more symmetric than $\mathcal{L}_c^k S^4$ toroidification $\mathcal{T}^k S^4$. This action universally represents symmetries of the equations of motion of supergravity in the reduction of M-theory to $11-k$ dimensions. Along the way, we identify the minimal model of the toroidification $\mathcal{T}^k S^4$, generalizing the results of Vigu\'{e}-Poirrier, Sullivan, and Burghelea, and establish an algebraic toroidification/totalization adjunction., Comment: 29 pages

Microgrids are evolving as promising options to enhance reliability of the connected transmission and distribution systems. Traditional design and deployment of microgrids require significant engineering analysis. Microgrid Building Blocks (MBB), consisting of modular blocks that integrate seamlessly to form effective microgrids, is an enabling concept for faster and broader adoption of microgrids. Back-to-Back converter placed at the point of common coupling of microgrid is an integral part of the MBB. This paper presents applications of MBB to decouple power quality issues in grid-microgrid network serving power quality sensitive loads such as data centers, new grid-edge technologies such as vehicle-to-grid generation, and serving electric vehicle charging loads during evacuation before disaster events. Simulation results show that MBB effectively decouples the power quality issues across networks and helps maintain good power quality in the power quality sensitive network based on the operational scenario., Comment: This paper is accepted for publication in IEEE IECON 2024, Chicago, IL. The complete copyright version will be available on IEEE Xplore when the conference proceedings are published

While fractional quantum Hall systems provide the best experimental evidence yet of (abelian) anyons plausibly necessary for future fault-tolerant quantum computation, like all strongly-coupled quantum systems their physics is not deeply understood. However, generally a promising approach is to (holographically) realize such systems on branes in string/M-theory; and specifically an old argument by Hellerman & Susskind gives a sketch of fractional quantum Hall states arising via discrete light cone quantization of M5/M9-brane intersections. Here we present a rigorous derivation of abelian anyon quantum states on M5$\perp$MO9-branes ("open M5-branes") on the discrete light cone, after globally completing the traditional local field content on the M5-worldvolume via a flux-quantization law compatible with the ambient 11d supergravity, specifically taken to be in the unstable co-Homotopy cohomology ("Hypothesis H"). The main step in the proof uses a theorem of Okuyama to identify co-Homotopy moduli spaces with configuration spaces of strings with charged endpoints, and identifies their loop spaces with cobordism of framed links that, under topological light cone quantization, turn out to be identified with the regularized Wilson loops of abelian Chern-Simons theory., Comment: 22 pages

Over a decade before the modern formulation of AdS/CFT duality, Duff et al. had observed a candidate microscopic explanation by identifying the CFT fields with fluctuations of probe p-branes stretched out in parallel near the horizon of their own black brane incarnation. A profound way to characterize these and more general probe p-brane configurations, especially for M5-branes, is expected to be as "super-embeddings" of their super-worldvolumes into target super-spacetime - but no concrete example of these had appeared in the literature. Here we fill this gap by constructing the explicit holographic super-embedding of probe M5-branes and M2-branes into their corresponding super-AdS backgrounds. The result seems to be novel and shows that the holographic M-probes, static with respect to the Poincar\'e chart, must sit exactly at the throat radius., Comment: 28 pages

Microgrids offer increased self-reliance and resilience at the grid's edge. They promote a significant transition to decentralized and renewable energy production by optimizing the utilization of local renewable sources. However, to maintain stable operations under all conditions and harness microgrids' full economic and technological potential, it is essential to integrate with the bulk grid and neighboring microgrids seamlessly. In this paper, we explore the capabilities of Back-to-Back (BTB) converters as a pivotal technology for interfacing microgrids, hybrid AC/DC grids, and bulk grids, by leveraging a comprehensive phasor-domain model integrated into GridLAB-D. The phasor-domain model is computationally efficient for simulating BTB with bulk grids and networked microgrids. We showcase the versatility of BTB converters (an integrated Microgrid Building Block) by configuring a two-microgrid network from a modified IEEE 13-node distribution system. These microgrids are equipped with diesel generators, photovoltaic units, and Battery Energy Storage Systems (BESS). The simulation studies are focused on use cases demonstrating dynamic decoupling and controlled support that a microgrid can provide via a BTB converter., Comment: This paper is accepted for publication in IEEE PES Grid Edge Technologies Conference & Exposition 2025, San Diego, CA. The complete copyright version will be available on IEEE Xplore when the conference proceedings are published

The bad science matrix problem consists in finding, among all matrices $A \in \mathbb{R}^{n \times n}$ with rows having unit $\ell^2$ norm, one that maximizes $\beta(A) = \frac{1}{2^n} \sum_{x \in \{-1, 1\}^n} \|Ax\|_\infty$. Our main contribution is an explicit construction of an $n \times n$ matrix $A$ showing that $\beta(A) \geq \sqrt{\log_2(n+1)}$, which is only 18% smaller than the asymptotic rate. We prove that every entry of any optimal matrix is a square root of a rational number, and we find provably optimal matrices for $n \leq 4$., Comment: 17 pages, 2 figures. Closest to version to be published in Involve, a Journal of Mathematics

The ubiquity of missing data has sparked considerable attention and focus on tabular data imputation methods. Diffusion models, recognized as the cutting-edge technique for data generation, demonstrate significant potential in tabular data imputation tasks. However, in pursuit of diversity, vanilla diffusion models often exhibit sensitivity to initialized noises, which hinders the models from generating stable and accurate imputation results. Additionally, the sparsity inherent in tabular data poses challenges for diffusion models in accurately modeling the data manifold, impacting the robustness of these models for data imputation. To tackle these challenges, this paper introduces an advanced diffusion model named Self-supervised imputation Diffusion Model (SimpDM for brevity), specifically tailored for tabular data imputation tasks. To mitigate sensitivity to noise, we introduce a self-supervised alignment mechanism that aims to regularize the model, ensuring consistent and stable imputation predictions. Furthermore, we introduce a carefully devised state-dependent data augmentation strategy within SimpDM, enhancing the robustness of the diffusion model when dealing with limited data. Extensive experiments demonstrate that SimpDM matches or outperforms state-of-the-art imputation methods across various scenarios., Comment: 10 pages, 5 figures. Accepted by CIKM 2024

We present ALLaM: Arabic Large Language Model, a series of large language models to support the ecosystem of Arabic Language Technologies (ALT). ALLaM is carefully trained considering the values of language alignment and knowledge transfer at scale. Our autoregressive decoder-only architecture models demonstrate how second-language acquisition via vocabulary expansion and pretraining on a mixture of Arabic and English text can steer a model towards a new language (Arabic) without any catastrophic forgetting in the original language (English). Furthermore, we highlight the effectiveness of using parallel/translated data to aid the process of knowledge alignment between languages. Finally, we show that extensive alignment with human preferences can significantly enhance the performance of a language model compared to models of a larger scale with lower quality alignment. ALLaM achieves state-of-the-art performance in various Arabic benchmarks, including MMLU Arabic, ACVA, and Arabic Exams. Our aligned models improve both in Arabic and English from their base aligned models.

The accurate evaluation of left atrial fibrosis via high-quality 3D Late Gadolinium Enhancement (LGE) MRI is crucial for atrial fibrillation management but is hindered by factors like patient movement and imaging variability. The pursuit of automated LGE MRI quality assessment is critical for enhancing diagnostic accuracy, standardizing evaluations, and improving patient outcomes. The deep learning models aimed at automating this process face significant challenges due to the scarcity of expert annotations, high computational costs, and the need to capture subtle diagnostic details in highly variable images. This study introduces HAMIL-QA, a multiple instance learning (MIL) framework, designed to overcome these obstacles. HAMIL-QA employs a hierarchical bag and sub-bag structure that allows for targeted analysis within sub-bags and aggregates insights at the volume level. This hierarchical MIL approach reduces reliance on extensive annotations, lessens computational load, and ensures clinically relevant quality predictions by focusing on diagnostically critical image features. Our experiments show that HAMIL-QA surpasses existing MIL methods and traditional supervised approaches in accuracy, AUROC, and F1-Score on an LGE MRI scan dataset, demonstrating its potential as a scalable solution for LGE MRI quality assessment automation. The code is available at: $\href{https://github.com/arf111/HAMIL-QA}{\text{this https URL}}$, Comment: Accepted to MICCAI2024, 10 pages, 2 figures

The integration of artificial intelligence (AI) chatbots into higher education marks a shift towards a new generation of pedagogical tools, mirroring the arrival of milestones like the internet. With the launch of ChatGPT-4 Turbo in November 2023, we developed a ChatGPT-based teaching application (https://chat.openai.com/g/g-1imx1py4K-chatge-medical-imaging) and integrated it into our undergraduate medical imaging course in the Spring 2024 semester. This study investigates the use of ChatGPT throughout a semester-long trial, providing insights into students' engagement, perception, and the overall educational effectiveness of the technology. We systematically collected and analyzed data concerning students' interaction with ChatGPT, focusing on their attitudes, concerns, and usage patterns. The findings indicate that ChatGPT offers significant advantages such as improved information access and increased interactivity, but its adoption is accompanied by concerns about the accuracy of the information provided and the necessity for well-defined guidelines to optimize its use.

Robust perception is critical for autonomous driving, especially under adverse weather and lighting conditions that commonly occur in real-world environments. In this paper, we introduce the Stereo Image Dataset (SID), a large-scale stereo-image dataset that captures a wide spectrum of challenging real-world environmental scenarios. Recorded at a rate of 20 Hz using a ZED stereo camera mounted on a vehicle, SID consists of 27 sequences totaling over 178k stereo image pairs that showcase conditions from clear skies to heavy snow, captured during the day, dusk, and night. The dataset includes detailed sequence-level annotations for weather conditions, time of day, location, and road conditions, along with instances of camera lens soiling, offering a realistic representation of the challenges in autonomous navigation. Our work aims to address a notable gap in research for autonomous driving systems by presenting high-fidelity stereo images essential for the development and testing of advanced perception algorithms. These algorithms support consistent and reliable operation across variable weather and lighting conditions, even when handling challenging situations like lens soiling. SID is publicly available at: https://doi.org/10.7302/esz6-nv83., Comment: This paper has been accepted for publication at the 2024 IEEE National Aerospace and Electronics Conference (NAECON), 2024

Accurate segmentation of medical images is crucial for diagnostic purposes, including cell segmentation, tumor identification, and organ localization. Traditional convolutional neural network (CNN)-based approaches struggled to achieve precise segmentation results due to their limited receptive fields, particularly in cases involving multi-organ segmentation with varying shapes and sizes. The transformer-based approaches address this limitation by leveraging the global receptive field, but they often face challenges in capturing local information required for pixel-precise segmentation. In this work, we introduce DwinFormer, a hierarchical encoder-decoder architecture for medical image segmentation comprising a directional window (Dwin) attention and global self-attention (GSA) for feature encoding. The focus of our design is the introduction of Dwin block within DwinFormer that effectively captures local and global information along the horizontal, vertical, and depthwise directions of the input feature map by separately performing attention in each of these directional volumes. To this end, our Dwin block introduces a nested Dwin attention (NDA) that progressively increases the receptive field in horizontal, vertical, and depthwise directions and a convolutional Dwin attention (CDA) that captures local contextual information for the attention computation. While the proposed Dwin block captures local and global dependencies at the first two high-resolution stages of DwinFormer, the GSA block encodes global dependencies at the last two lower-resolution stages. Experiments over the challenging 3D Synapse Multi-organ dataset and Cell HMS dataset demonstrate the benefits of our DwinFormer over the state-of-the-art approaches. Our source code will be publicly available at \url{https://github.com/Daniyanaj/DWINFORMER}., Comment: 5 pages

We highlight the need for global completion of the field content in the M5-brane sigma-model analogous to Dirac's charge/flux quantization, and we point out that the superspace Bianchi identities on the worldvolume and on its ambient supergravity background constrain the M5's flux-quantization law to be in a non-abelian cohomology theory rationally equivalent to a twisted form of co-Homotopy. In order to clearly bring out this subtle point we give a streamlined re-derivation of the worldvolume 3-flux via M5 "super-embeddings". Finally, assuming the flux-quantization law to actually be in co-Homotopy ("Hypothesis H") we show how this implies Skyrmion-like solitons on general M5-worldvolumes and (abelian) anyonic solitons on the boundaries of "open M5-branes" in heterotic M-theory., Comment: 44 pages; v3: published version, abstract shortened and typos fixed; v2: minor additions

Drawing upon StyleGAN's expressivity and disentangled latent space, existing 2D approaches employ textual prompting to edit facial images with different attributes. In contrast, 3D-aware approaches that generate faces at different target poses require attribute-specific classifiers, learning separate model weights for each attribute, and are not scalable for novel attributes. In this work, we propose an efficient, plug-and-play, 3D-aware face editing framework based on attribute-specific prompt learning, enabling the generation of facial images with controllable attributes across various target poses. To this end, we introduce a text-driven learnable style token-based latent attribute editor (LAE). The LAE harnesses a pre-trained vision-language model to find text-guided attribute-specific editing direction in the latent space of any pre-trained 3D-aware GAN. It utilizes learnable style tokens and style mappers to learn and transform this editing direction to 3D latent space. To train LAE with multiple attributes, we use directional contrastive loss and style token loss. Furthermore, to ensure view consistency and identity preservation across different poses and attributes, we employ several 3D-aware identity and pose preservation losses. Our experiments show that our proposed framework generates high-quality images with 3D awareness and view consistency while maintaining attribute-specific features. We demonstrate the effectiveness of our method on different facial attributes, including hair color and style, expression, and others., Comment: Accepted at ECCV, 2024. Amandeep Kumar and Muhammad Awais are joint first authors. More details are available at https://awaisrauf.github.io/3d_face_editing

Recent works on open-vocabulary 3D instance segmentation show strong promise, but at the cost of slow inference speed and high computation requirements. This high computation cost is typically due to their heavy reliance on 3D clip features, which require computationally expensive 2D foundation models like Segment Anything (SAM) and CLIP for multi-view aggregation into 3D. As a consequence, this hampers their applicability in many real-world applications that require both fast and accurate predictions. To this end, we propose a fast yet accurate open-vocabulary 3D instance segmentation approach, named Open-YOLO 3D, that effectively leverages only 2D object detection from multi-view RGB images for open-vocabulary 3D instance segmentation. We address this task by generating class-agnostic 3D masks for objects in the scene and associating them with text prompts. We observe that the projection of class-agnostic 3D point cloud instances already holds instance information; thus, using SAM might only result in redundancy that unnecessarily increases the inference time. We empirically find that a better performance of matching text prompts to 3D masks can be achieved in a faster fashion with a 2D object detector. We validate our Open-YOLO 3D on two benchmarks, ScanNet200 and Replica, under two scenarios: (i) with ground truth masks, where labels are required for given object proposals, and (ii) with class-agnostic 3D proposals generated from a 3D proposal network. Our Open-YOLO 3D achieves state-of-the-art performance on both datasets while obtaining up to $\sim$16$\times$ speedup compared to the best existing method in literature. On ScanNet200 val. set, our Open-YOLO 3D achieves mean average precision (mAP) of 24.7\% while operating at 22 seconds per scene. Code and model are available at github.com/aminebdj/OpenYOLO3D.