Search

Flexoelectricity, a coupling between strain gradients and electric polarization, has attracted significant interest due to its critical role in enhanced effects at small scales and its applicability across a diverse range of materials. Modeling flexoelectricity is challenging, especially in 3D, due to the higher-order equations involved, which require continuity conditions that limit the use of standard finite element methods. This study employs a modified mixed finite element formulation, specifically designed to account for spurious oscillations that can hinder convergence, particularly in large-scale problems solved with iterative solvers. A numerical analysis has been conducted to evaluate the effectiveness of stacking individual flexoelectric unit cells to achieve an enhanced overall flexoelectric response. The study also seeks to determine the feasibility of locally modifying the strain gradient to induce localized enhancements in the flexoelectric effect.

Exploring the topological characteristics of electronic bands is essential in condensed matter physics. Moir\'e materials featuring flat bands provide a versatile platform for engineering band topology and correlation effects. In moir\'e materials that break either time-reversal symmetry or inversion symmetry or both, electronic bands exhibit Berry curvature hotspots. Different stacking orders in these materials result in varied Berry curvature distributions within the flat bands, even when the band dispersion remains similar. However, experimental studies probing the impact of stacking order on the quantum geometric quantities are lacking. 1.4$^\circ$ twisted double bilayer graphene (TDBG) facilitates two distinct stacking orders (AB-AB, AB-BA) and forms an inversion broken \moire superlattice with electrically tunable flat bands. The valley Chern numbers of the flat bands depend on the stacking order, and the nonlinear Hall (NLH) effect distinguishes the differences in Berry curvature dipole (BCD), the first moment of Berry curvature. The BCD exhibits antisymmetric behavior, flipping its sign with the polarity of the perpendicular electric field in AB-AB TDBG, while it displays a symmetric behavior, maintaining the same sign regardless of the electric field's polarity in AB-BA TDBG. This approach electronically detects stacking-induced quantum geometry, while opening a pathway to quantum geometry engineering and detection.

A python library for the creation of MCNP6 input decks is described. The library supports geometry generation with automatic assignment of surface/facet numbers, cell numbers, transform numbers and material numbers along with MCNP Universes and FILL feature. Rectangular and Hexagonal Lattices are also supported. A large material library is included. Support for a good selection of common sources and tallies is also provided. Cards or features which are currently not supported in the library can also be inserted as raw strings into the output stream. Combining Python features like descriptively named variables, functions and for loops with library functions provides an intuitive and parametric way to create, modify and maintain complicated geometries and simulation models. The generated card deck also has human readable comments which makes it easy to read and relate back to the python source. Some support for running MCNP, reading tallies and plotting is also provided., Comment: 12 pages, 3 figures

In this article we study the $L^p$-improving mapping properties of the totally-geodesic $k$-plane transform on simply connected spaces of constant curvature, namely, $\mathbb{R}^n$, $\mathbb{H}^n$ and $\mathbb{S}^n$. We begin our study by answering the question of the existence of the totally-geodesic $k$-plane transform on weighted $L^p$ spaces {with radial weights arising from the volume growth on these spaces}. These weights arise naturally from the geometry of these spaces. We then derive necessary and sufficient conditions for the $k$-plane transform of radial functions to be bounded on weighted Lebesgue spaces, with radial power weights. {Following an idea of Kurusa,} we also {derive} formulae for the totally-geodesic $k$-plane transform of general functions on the hyperbolic space and the sphere. Using this formula, and an elementary technique of Minkowski inequality, we prove weighted $L^p$-$L^p$ boundedness of the $k$-plane transform of general functions as well. Along with this, we also study the end-point behaviour of the transform, where the ``end-point" naturally arises due to either the existence conditions or the necessary conditions for boundedness., Comment: 83 pages, 7 figures

Highly sensitive, broadband bolometers are of great interest because of their versatile usage in wide areas starting from dark matter search, radio astronomy, material science, and qubit readouts in cQED experiments. There have been different realizations of bolometers using superconducting thin films, nanowires, quantum dots, and various 2D materials in the recent past. The challenge is to have a single device that combines high sensitivity, broad bandwidth, a fast readout mechanism, and low noise. Here we demonstrate the first usage of a Josephson parametric amplifier (JPA) as a highly sensitive bolometer. Our key finding is the Kerr non-linearity of the JPA boosts the device's sensitivity. When the bolometer is biased in the non-linear regime, it enhances the sideband signals (~100 times), resulting in an order of magnitude improvement in sensitivity compared to the linear regime. In the non-linear biasing of the device, we achieve a NEP~500 aW/sqrt(Hz). Our bolometer offers a fast detection scheme with a thermal time constant of 4.26 us and an intrinsic JPA time constant of 70 ns. Our device's broadband and fast operation are key and new compared to previously studied graphene-based bolometers. In our device, the gate voltage tunability and the possibility of multiplexing combined with the sensitive bolometric performance offer an opportunity for integrated quantum sensor arrays. Our work demonstrates a way forward to enhance the performance of quantum sensors based on 2D materials by leveraging the inherent non-linear response.

Understanding and explaining differences between audio recordings is crucial for fields like audio forensics, quality assessment, and audio generation. This involves identifying and describing audio events, acoustic scenes, signal characteristics, and their emotional impact on listeners. This paper stands out as the first work to comprehensively study the task of explaining audio differences and then propose benchmark, baselines for the task. First, we present two new datasets for audio difference explanation derived from the AudioCaps and Clotho audio captioning datasets. Using Large Language Models (LLMs), we generate three levels of difference explanations: (1) concise descriptions of audio events and objects, (2) brief sentences about audio events, acoustic scenes, and signal properties, and (3) comprehensive explanations that include semantics and listener emotions. For the baseline, we use prefix tuning where audio embeddings from two audio files are used to prompt a frozen language model. Our empirical analysis and ablation studies reveal that the naive baseline struggles to distinguish perceptually similar sounds and generate detailed tier 3 explanations. To address these limitations, we propose ADIFF, which introduces a cross-projection module, position captioning, and a three-step training process to enhance the model's ability to produce detailed explanations. We evaluate our model using objective metrics and human evaluation and show our model enhancements lead to significant improvements in performance over naive baseline and SoTA Audio-Language Model (ALM) Qwen Audio. Lastly, we conduct multiple ablation studies to study the effects of cross-projection, language model parameters, position captioning, third stage fine-tuning, and present our findings. Our benchmarks, findings, and strong baseline pave the way for nuanced and human-like explanations of audio differences., Comment: Accepted at ICLR 2025. Dataset and checkpoints are available at: https://github.com/soham97/ADIFF

Optical Character Recognition (OCR) technology is widely used to extract text from images of documents, facilitating efficient digitization and data retrieval. However, merely extracting text is insufficient when dealing with complex documents. Fully comprehending such documents requires an understanding of their structure -- including formatting, formulas, tables, and the reading order of multiple blocks and columns across multiple pages -- as well as semantic information for detecting elements like footnotes and image captions. This comprehensive understanding is crucial for downstream tasks such as retrieval, document question answering, and data curation for training Large Language Models (LLMs) and Vision Language Models (VLMs). To address this, we introduce \'Eclair, a general-purpose text-extraction tool specifically designed to process a wide range of document types. Given an image, \'Eclair is able to extract formatted text in reading order, along with bounding boxes and their corresponding semantic classes. To thoroughly evaluate these novel capabilities, we introduce our diverse human-annotated benchmark for document-level OCR and semantic classification. \'Eclair achieves state-of-the-art accuracy on this benchmark, outperforming other methods across key metrics. Additionally, we evaluate \'Eclair on established benchmarks, demonstrating its versatility and strength across several evaluation standards.

Given the uncertain evolutionary status of blue supergiant stars, their multiplicity properties hold vital clues to better understand their origin and evolution. As part of The Binarity at LOw Metallicity (BLOeM) campaign in the Small Magellanic Cloud we present a multi-epoch spectroscopic survey of 128 supergiant stars of spectral type B5--F5, which roughly correspond to initial masses in the range 6 to 30 solar masses. The observed binary fraction for the B5-9 supergiants is 25+/-6 % (10+/-4 %) and 5+/-2 % (0 %) for the A-F stars, using a radial velocity (RV) variability threshold of 5 kms (10 kms) as a criterion for binarity. Accounting for observational biases we find an intrinsic multiplicity fraction of less than 18% for the B5-9 stars and 8$^{+9}_{-7}$% for the AF stars, for the orbital periods up to 10$^{3.5}$day and mass-ratios (q) in the range 0.1 < q < 1. The large stellar radii of these supergiant stars prevent short orbital periods but we demonstrate that this effect alone cannot explain our results. We assess the spectra and RV time series of the detected binary systems and find that only a small fraction display convincing solutions. We conclude that the multiplicity fractions are compromised by intrinsic stellar variability such that the true multiplicity fraction may be significantly smaller. Our main conclusions from comparing the multiplicity properties of the B5-9 and AF supergiants to that of their less evolved counterparts is that such stars cannot be explained by a direct evolution from the main sequence. Furthermore, by comparing their multiplicity properties to red supergiant stars we conclude that the AF supergiant stars are neither progenitors nor descendants of red supergiants., Comment: Accepted in A&A

Rapidly rotating classical OBe stars have been proposed as the products of binary interactions, and the fraction of Be stars with compact companions implies that at least some are. However, to constrain the interaction physics spinning up the OBe stars, a large sample of homogeneously analysed OBe stars with well-determined binary characteristics and orbital parameters are required. We investigate the multiplicity properties of a sample of 18 Oe, 62 Be, and two Of?p stars observed within the BLOeM survey in the Small Magellanic Cloud. We analyse the first nine epochs of spectroscopic observations obtained over approximately three months in 2023. Radial velocities (RVs) of all stars are measured. Applying commonly-used binarity criteria we classify objects as binaries, binary candidates, and apparently single (RV stable) objects. We further inspect the spectra for double-lined spectroscopic binaries and cross-match with catalogues of X-ray sources and photometric binaries. We classify 14 OBe stars as binaries, and an additional 11 as binary candidates. The two Of?p stars are apparently single. Two more objects are most likely currently interacting binaries. Without those, the observed binary fraction for the OBe sample (78 stars) is f_OBe_obs=0.18+/-0.04 (f_obs_cand=0.32+/-0.05 including candidates). This fraction is less than half of that measured for OB stars in BLOeM. Combined with the lower fraction of SB2s, this suggests that OBe stars have indeed fundamentally different binary properties than OB stars. We find no evidence for OBe binaries with massive compact companions, in contrast to expectations from binary population synthesis. Our results support the binary scenario as an important formation channel for OBe stars, as post-interaction binaries may have been disrupted or the stripped companions of OBe stars are harder to detect., Comment: 12 pages (+appendix), 7 figures, accepted for publication in A&A

Accurate estimation of human hand configuration and the forces they exert is critical for effective teleoperation and skill transfer in robotic manipulation. A deeper understanding of human interactions with objects can further enhance teleoperation performance. To address this need, researchers have explored methods to capture and translate human manipulation skills and applied forces to robotic systems. Among these, biosignal-based approaches, particularly those using forearm ultrasound data, have shown significant potential for estimating hand movements and finger forces. In this study, we present a method for simultaneously estimating manipulation skills and applied hand force using forearm ultrasound data. Data collected from seven participants were used to train deep learning models for classifying manipulation skills and estimating grasp force. Our models achieved an average classification accuracy of 94.87 percent plus or minus 10.16 percent for manipulation skills and an average root mean square error (RMSE) of 0.51 plus or minus 0.19 Newtons for force estimation, as evaluated using five-fold cross-validation. These results highlight the effectiveness of forearm ultrasound in advancing human-machine interfacing and robotic teleoperation for complex manipulation tasks. This work enables new and effective possibilities for human-robot skill transfer and tele-manipulation, bridging the gap between human dexterity and robotic control., Comment: 30 pages, 52 references, 10 figures, 8 tables and 2 supplementary videos. Currently under review

The measurement of a superconducting qubit is implemented by coupling it to a resonator. The common choice is transverse coupling, which, in the dispersive approximation, introduces an interaction term which enables the measurement. This cross-Kerr term provides a qubit-state dependent dispersive shift in the resonator frequency with the device parameters chosen carefully to get sufficient signal while minimizing Purcell decay of the qubit. We introduce a two-mode circuit, nicknamed quantromon, with two orthogonal modes implementing a qubit and a resonator. Unlike before, where the coupling term emerges as a perturbative expansion, the quantromon has intrinsic cross-Kerr coupling by design. Our experiments implemented in a hybrid 2D-3D cQED architecture demonstrate some unique features of the quantromon like weak dependence of the dispersive shift on the qubit-resonator detuning and intrinsic Purcell protection. In a tunable qubit-frequency device, we show that the dispersive shift ($2\chi/2\pi$) changes by only $0.8$ MHz while the qubit-resonator detuning ($\Delta/2\pi$) is varied between $0.398$ GHz - $3.288$ GHz. We also demonstrate Purcell protection in a second device where we tune the orthogonality between the two modes. Finally, we demonstrate a single-shot readout fidelity of $98.3\%$ without using a parametric amplifier which is comparable to the state-of-the-art and suggests a potential simplification of the measurement circuitry for scaling up quantum processors., Comment: Manuscript (6 pages, 4 figures), Supplement (4 pages, 2 figures)

Implementing a precise measurement of the quantum state of a qubit is very critical for building a practical quantum processor as it plays an important role in state initialization and quantum error correction. While the transmon qubit has been the most commonly used design in small to medium-scale processors, the fluxonium qubit is emerging as a strong alternative with the potential for high-fidelity gate operation as a result of the high anharmonicity and high coherence achievable due to its unique design. Here, we explore the measurement characteristics of a tantalum-based high-coherence fluxonium qubit and demonstrate single-shot measurement fidelity (assignment fidelity) of 96.2% and 97.8% without and with the use of a Josephson Parametric Amplifier respectively. We study the back-action of the measurement photons on the qubit and measure a QND (repeatability) fidelity of 99.6%. We find that the measurement fidelity and QND nature are limited by state-mixing errors and our results suggest that a careful study of measurement-induced transitions in the fluxonium is needed to further optimize the readout performance., Comment: 9 pages, 9 figures

Recently, promising progress has been made by open-source vision-language models (VLMs) in bringing their capabilities closer to those of proprietary frontier models. However, most open-source models only publish their final model weights, leaving the critical details of data strategies and implementation largely opaque. In this work, we address VLM post-training from a data-centric perspective, showing the key role of data strategy in developing frontier VLMs. By studying and building our post-training data strategy from scratch, we share detailed insights into the development processes, aiming to benefit the development of competitive models for the open-source community. Our introduced data strategy, together with training recipes and model design, leads to a family of performant VLMs named Eagle2. Specifically, Eagle2-9B achieves state-of-the-art results across various multimodal benchmarks, matching certain competitive models with up to 70B parameters.

A next-generation medium-energy gamma-ray telescope targeting the MeV range would address open questions in astrophysics regarding how extreme conditions accelerate cosmic-ray particles, produce relativistic jet outflows, and more. One concept, AMEGO-X, relies upon the mission-enabling CMOS Monolithic Active Pixel Sensor silicon chip AstroPix. AstroPix is designed for space-based use, featuring low noise, low power consumption, and high scalability. Desired performance of the device include an energy resolution of 5 keV (or 10% FWHM) at 122 keV and a dynamic range per-pixel of 25-700 keV, enabled by the addition of a high-voltage bias to each pixel which supports a depletion depth of 500 um. This work reports on the status of the AstroPix development process with emphasis on the current version under test, version three (v3), and highlights of version two (v2). Version 3 achieves energy resolution of 10.4 +\- 3.2 % at 59.5 keV and 94 +\- 6 um depletion in a low-resistivity test silicon substrate., Comment: 20 pages, 13 figures

As gravitational wave detectors become more advanced and sensitive, the number of signals recorded by Advanced LIGO and Virgo from merging compact objects is expected to rise dramatically. This surge in detection rates necessitates the development of adaptable, scalable, and efficient tools capable of addressing a wide range of tasks in gravitational wave astronomy. Foundational AI models present a transformative opportunity in this context by providing a unified framework that can be fine tuned for diverse applications while leveraging the power of large scale pre training. In this work, we explore how advanced transformer models, specifically Whisper by OpenAI, can be adapted as a foundational model for gravitational wave data analysis. By fine tuning the encoder model of Whisper, originally trained on extensive audio data, and combining it with neural networks for specialized tasks, we achieve reliable results in detecting astrophysical signals and classifying transient noise artifacts or glitches. This represents the first application of open source transformer models, pre trained on unrelated tasks, for gravitational wave research, demonstrating their potential to enable versatile and efficient data analysis in the era of rapidly increasing detection rates.

Small droplets are irradiated with visible and infrared light in many natural and industrial environments. One of the simplest ways to describe their evaporation is the D$^2$-Law. It states that the evaporation rate is proportional to $t^{-1/2}$, and $R^{-1}$. However, models like the D$^2$-Law do not account for the volumetric heating of light and the effect of strong irradiation on individual droplets is not fully understood. Here we show the effects of IR irradiation on optically levitated water droplets. We find that, under strong irradiation of up to $10^8 W/m^2$, the droplet evaporation is initially driven by the heat from the laser following the power law $dR / dt \sim R$, i.e. the inverse of the D$^2$-Law. Then, when the droplets shrink to 2 - 3 $\mu$m in radius a turnover occurs from irradiation-driven back to diffusion-driven evaporation. Our findings support the understanding of droplet evaporation in cases such as rocket engines or internal combustion, where the radiation from the flame will heat water and fuel droplets., Comment: 9 pages, 5 figures

The global outbreak of the Mpox virus, classified as a Public Health Emergency of International Concern (PHEIC) by the World Health Organization, presents significant diagnostic challenges due to its visual similarity to other skin lesion diseases. Traditional diagnostic methods for Mpox, which rely on clinical symptoms and laboratory tests, are slow and labor intensive. Deep learning-based approaches for skin lesion classification offer a promising alternative. However, developing a model that balances efficiency with accuracy is crucial to ensure reliable and timely diagnosis without compromising performance. This study introduces the Cascaded Atrous Group Attention (CAGA) framework to address these challenges, combining the Cascaded Atrous Attention module and the Cascaded Group Attention mechanism. The Cascaded Atrous Attention module utilizes dilated convolutions and cascades the outputs to enhance multi-scale representation. This is integrated into the Cascaded Group Attention mechanism, which reduces redundancy in Multi-Head Self-Attention. By integrating the Cascaded Atrous Group Attention module with EfficientViT-L1 as the backbone architecture, this approach achieves state-of-the-art performance, reaching an accuracy of 98% on the Mpox Close Skin Image (MCSI) dataset while reducing model parameters by 37.5% compared to the original EfficientViT-L1. The model's robustness is demonstrated through extensive validation on two additional benchmark datasets, where it consistently outperforms existing approaches., Comment: 8 pages, 4 figures, Submitted to Medical Imaging with Deep Learning

The hearing-impaired community in India deserves the access to tools that help them communicate, however, there is limited known technology solutions that make use of Indian Sign Language (ISL) at present. Even though there are many ISL users, ISL cannot access social and education arenas because there is not yet an efficient technology to convert the ISL signal into speech or text. We initiated this initiative owing to the rising demand for products and technologies that are inclusive and help ISL, filling the gap of communication for the ones with hearing disability. Our goal is to build an reliable sign language recognition system with the help of Convolutional Neural Networks (CNN) to . By expanding communication access, we aspire toward better educational opportunities and a more inclusive society for hearing impaired people in India.

Retrieval-Augmented Generation (RAG) enhances Large Language Models (LLMs) by incorporating external, domain-specific data into the generative process. While LLMs are highly capable, they often rely on static, pre-trained datasets, limiting their ability to integrate dynamic or private data. Traditional RAG systems typically use a single-agent architecture to handle query generation, data retrieval, and response synthesis. However, this approach becomes inefficient when dealing with diverse data sources, such as relational databases, document stores, and graph databases, often leading to performance bottlenecks and reduced accuracy. This paper proposes a multi-agent RAG system to address these limitations. Specialized agents, each optimized for a specific data source, handle query generation for relational, NoSQL, and document-based systems. These agents collaborate within a modular framework, with query execution delegated to an environment designed for compatibility across various database types. This distributed approach enhances query efficiency, reduces token overhead, and improves response accuracy by ensuring that each agent focuses on its specialized task. The proposed system is scalable and adaptable, making it ideal for generative AI workflows that require integration with diverse, dynamic, or private data sources. By leveraging specialized agents and a modular execution environment, the system provides an efficient and robust solution for handling complex, heterogeneous data environments in generative AI applications., Comment: 16 pages, 3 figures. This preprint introduces a multi-agent framework for Retrieval-Augmented Generation (RAG), enhancing Large Language Models (LLMs) for efficient integration of diverse data sources. Relevant for researchers in AI, ML, generative AI, and database systems

Multi-objective alignment from human feedback (MOAHF) in large language models (LLMs) is a challenging problem as human preferences are complex, multifaceted, and often conflicting. Recent works on MOAHF considered a-priori multi-objective optimization (MOO), where human preferences are known at training or inference time. In contrast, when human preferences are unknown or difficult to quantify, a natural approach is to cover the Pareto front by multiple diverse solutions. We propose an algorithm HaM for learning diverse LLM policies that maximizes their hypervolume. This is the first application of a-posteriori MOO to MOAHF. HaM is computationally and space efficient, and empirically superior across objectives such as harmlessness, helpfulness, humor, faithfulness, and hallucination, on various datasets.

We develop a motivic cohomology theory, representable in the Voevodsky's triangulated category of motives, for smooth separated Deligne-Mumford stacks and show that the resulting higher Chow groups are canonically isomorphic to the higher $K$-theory of such stacks. This generalises the Grothendieck-Riemann-Roch theorem to the category of smooth Deligne-Mumford stacks., Comment: 10 pages. Comments welcome

This report presents a comprehensive response to the United Nation's Interim Report on Governing Artificial Intelligence (AI) for Humanity. It emphasizes the transformative potential of AI in achieving the Sustainable Development Goals (SDGs) while acknowledging the need for robust governance to mitigate associated risks. The response highlights opportunities for promoting equitable, secure, and inclusive AI ecosystems, which should be supported by investments in infrastructure and multi-stakeholder collaborations across jurisdictions. It also underscores challenges, including societal inequalities exacerbated by AI, ethical concerns, and environmental impacts. Recommendations advocate for legally binding norms, transparency, and multi-layered data governance models, alongside fostering AI literacy and capacity-building initiatives. Internationally, the report calls for harmonising AI governance frameworks with established laws, human rights standards, and regulatory approaches. The report concludes with actionable principles for fostering responsible AI governance through collaboration among governments, industry, academia, and civil society, ensuring the development of AI aligns with universal human values and the public good., Comment: Submitted to United Nations. 23 pages. All the Authors Contributed Equally

This paper explores the application of a reinforcement learning (RL) framework using the Q-Learning algorithm to enhance dynamic pricing strategies in the retail sector. Unlike traditional pricing methods, which often rely on static demand models, our RL approach continuously adapts to evolving market dynamics, offering a more flexible and responsive pricing strategy. By creating a simulated retail environment, we demonstrate how RL effectively addresses real-time changes in consumer behavior and market conditions, leading to improved revenue outcomes. Our results illustrate that the RL model not only surpasses traditional methods in terms of revenue generation but also provides insights into the complex interplay of price elasticity and consumer demand. This research underlines the significant potential of applying artificial intelligence in economic decision-making, paving the way for more sophisticated, data-driven pricing models in various commercial domains., Comment: This paper has been accepted for presentation at the 1st IEEE International Conference on AIML-Applications for Engineering & Technology (ICAET-25)

The newly introduced Visual State Space Model (VMamba), which employs \textit{State Space Mechanisms} (SSM) to interpret images as sequences of patches, has shown exceptional performance compared to Vision Transformers (ViT) across various computer vision tasks. However, recent studies have highlighted that deep models are susceptible to adversarial attacks. One common approach is to embed a trigger in the training data to retrain the model, causing it to misclassify data samples into a target class, a phenomenon known as a backdoor attack. In this paper, we first evaluate the robustness of the VMamba model against existing backdoor attacks. Based on this evaluation, we introduce a novel architectural backdoor attack, termed BadScan, designed to deceive the VMamba model. This attack utilizes bit plane slicing to create visually imperceptible backdoored images. During testing, if a trigger is detected by performing XOR operations between the $k^{th}$ bit planes of the modified triggered patches, the traditional 2D selective scan (SS2D) mechanism in the visual state space (VSS) block of VMamba is replaced with our newly designed BadScan block, which incorporates four newly developed scanning patterns. We demonstrate that the BadScan backdoor attack represents a significant threat to visual state space models and remains effective even after complete retraining from scratch. Experimental results on two widely used image classification datasets, CIFAR-10, and ImageNet-1K, reveal that while visual state space models generally exhibit robustness against current backdoor attacks, the BadScan attack is particularly effective, achieving a higher Triggered Accuracy Ratio (TAR) in misleading the VMamba model and its variants.

Specialized datasets that capture the fashion industry's rich language and styling elements can boost progress in AI-driven fashion design. We present FLORA (Fashion Language Outfit Representation for Apparel Generation), the first comprehensive dataset containing 4,330 curated pairs of fashion outfits and corresponding textual descriptions. Each description utilizes industry-specific terminology and jargon commonly used by professional fashion designers, providing precise and detailed insights into the outfits. Hence, the dataset captures the delicate features and subtle stylistic elements necessary to create high-fidelity fashion designs. We demonstrate that fine-tuning generative models on the FLORA dataset significantly enhances their capability to generate accurate and stylistically rich images from textual descriptions of fashion sketches. FLORA will catalyze the creation of advanced AI models capable of comprehending and producing subtle, stylistically rich fashion designs. It will also help fashion designers and end-users to bring their ideas to life. As a second orthogonal contribution, we introduce KAN Adapters, which leverage Kolmogorov-Arnold Networks (KAN) as adaptive modules. They serve as replacements for traditional MLP-based LoRA adapters. With learnable spline-based activations, KAN Adapters excel in modeling complex, non-linear relationships, achieving superior fidelity, faster convergence and semantic alignment. Extensive experiments and ablation studies on our proposed FLORA dataset validate the superiority of KAN Adapters over LoRA adapters. To foster further research and collaboration, we will open-source both the FLORA and our implementation code., Comment: Under review at a conference

Autonomous multi-agent systems such as hospital robots and package delivery drones often operate in highly uncertain environments and are expected to achieve complex temporal task objectives while ensuring safety. While learning-based methods such as reinforcement learning are popular methods to train single and multi-agent autonomous systems under user-specified and state-based reward functions, applying these methods to satisfy trajectory-level task objectives is a challenging problem. Our first contribution is the use of weighted automata to specify trajectory-level objectives, such that, maximal paths induced in the weighted automaton correspond to desired trajectory-level behaviors. We show how weighted automata-based specifications go beyond timeliness properties focused on deadlines to performance properties such as expeditiousness. Our second contribution is the use of evolutionary game theory (EGT) principles to train homogeneous multi-agent teams targeting homogeneous task objectives. We show how shared experiences of agents and EGT-based policy updates allow us to outperform state-of-the-art reinforcement learning (RL) methods in minimizing path length by nearly 30\% in large spaces. We also show that our algorithm is computationally faster than deep RL methods by at least an order of magnitude. Additionally our results indicate that it scales better with an increase in the number of agents as compared to other methods.

Let $R$ be a commutative ring spectrum. We construct the wrapped Donaldson--Fukaya category with coefficients in $R$ of any stably polarized Liouville sector. We show that any two $R$-orientable and isomorphic objects admit $R$-orientations so that their $R$-fundamental classes coincide. Our main result is that any closed exact Lagrangian $R$-brane in the cotangent bundle of a closed manifold is isomorphic to an $R$-brane structure on the zero section in the wrapped Donaldson--Fukaya category, generalizing a well-known result over the integers. To achieve this, we prove that the Floer homotopy type of the cotangent fiber is given by the stable homotopy type of the based loop space of the zero section., Comment: 99 pages, no figures. Comments welcome

Nonlinear interactions between single quantum particles are at the heart of any quantum information system, including analog quantum simulation and fault-tolerant quantum computing. This remains a particularly difficult problem for photonic qubits, as photons do not interact with each other. While engineering light-matter interaction can effectively create photon-photon interaction, the required photon number to observe any nonlinearity is very high, where any quantum mechanical signature disappears. However, with emerging low-dimensional materials, and engineered photonic resonators, the photon number can be potentially reduced to reach the quantum nonlinear optical regime. In this review paper, we discuss different mechanisms exploited in solid-state platforms to attain quantum nonlinear optics. We review emerging materials and optical resonator architecture with different dimensionalities. We also present new research directions and open problems in this field.

The Automated Audio Captioning (AAC) task aims to describe an audio signal using natural language. To evaluate machine-generated captions, the metrics should take into account audio events, acoustic scenes, paralinguistics, signal characteristics, and other audio information. Traditional AAC evaluation relies on natural language generation metrics like ROUGE and BLEU, image captioning metrics such as SPICE and CIDEr, or Sentence-BERT embedding similarity. However, these metrics only compare generated captions to human references, overlooking the audio signal itself. In this work, we propose MACE (Multimodal Audio-Caption Evaluation), a novel metric that integrates both audio and reference captions for comprehensive audio caption evaluation. MACE incorporates audio information from audio as well as predicted and reference captions and weights it with a fluency penalty. Our experiments demonstrate MACE's superior performance in predicting human quality judgments compared to traditional metrics. Specifically, MACE achieves a 3.28% and 4.36% relative accuracy improvement over the FENSE metric on the AudioCaps-Eval and Clotho-Eval datasets respectively. Moreover, it significantly outperforms all the previous metrics on the audio captioning evaluation task. The metric is opensourced at https://github.com/satvik-dixit/mace

Federated Learning (FL) paradigms enable large numbers of clients to collaboratively train Machine Learning models on private data. However, due to their multi-party nature, traditional FL schemes are left vulnerable to Byzantine attacks that attempt to hurt model performance by injecting malicious backdoors. A wide variety of prevention methods have been proposed to protect frameworks from such attacks. This paper provides a exhaustive and updated taxonomy of existing methods and frameworks, before zooming in and conducting an in-depth analysis of the strengths and weaknesses of the Robustness of Federated Learning (RoFL) protocol. From there, we propose two novel Sybil-based attacks that take advantage of vulnerabilities in RoFL. Finally, we conclude with comprehensive proposals for future testing, describe and detail implementation of the proposed attacks, and offer direction for improvements in the RoFL protocol as well as Byzantine-robust frameworks as a whole., Comment: 16 pages, 4 figures, 1 appendix

Green hydrogen has the potential to address two pressing problems in a zero-carbon energy system: balancing seasonal variability of solar and wind in the electricity sector, and replacing fossil fuels in hard-to-abate sectors. However, the previous research only separately modeled the electricity and hard-to-abate sectors, which is unable to capture how the interaction between the two sectors influences the energy system cost. In this study, focusing on China, we deploy an electricity system planning model to examine the cost implications of green hydrogen to fully decarbonize the electricity system and hard-to-abate sectors. Our results reveal that green hydrogen enables a 17% reduction in the levelized cost of a zero-carbon electricity system relative to that without hydrogen. However, cost savings hinge on the availability of underground hydrogen storage capacities and electric transmission expansion. More importantly, coupling hydrogen infrastructure in the electricity and hard-to-abate sectors not only reduces energy costs compared to a decoupled energy system but also makes green hydrogen cost-competitive compared to fossil fuel-based gray and blue hydrogen in China., Comment: 4 figures, 33 pages

Reinforcement learning (RL) has been successfully applied to solve the problem of finding obstacle-free paths for autonomous agents operating in stochastic and uncertain environments. However, when the underlying stochastic dynamics of the environment experiences drastic distribution shifts, the optimal policy obtained in the trained environment may be sub-optimal or may entirely fail in helping find goal-reaching paths for the agent. Approaches like domain randomization and robust RL can provide robust policies, but typically assume minor (bounded) distribution shifts. For substantial distribution shifts, retraining (either with a warm-start policy or from scratch) is an alternative approach. In this paper, we develop a novel approach called {\em Evolutionary Robust Policy Optimization} (ERPO), an adaptive re-training algorithm inspired by evolutionary game theory (EGT). ERPO learns an optimal policy for the shifted environment iteratively using a temperature parameter that controls the trade off between exploration and adherence to the old optimal policy. The policy update itself is an instantiation of the replicator dynamics used in EGT. We show that under fairly common sparsity assumptions on rewards in such environments, ERPO converges to the optimal policy in the shifted environment. We empirically demonstrate that for path finding tasks in a number of environments, ERPO outperforms several popular RL and deep RL algorithms (PPO, A3C, DQN) in many scenarios and popular environments. This includes scenarios where the RL algorithms are allowed to train from scratch in the new environment, when they are retrained on the new environment, or when they are used in conjunction with domain randomization. ERPO shows faster policy adaptation, higher average rewards, and reduced computational costs in policy adaptation., Comment: Pubblished in ECAI 2024

Pre-trained language models (PLMs) are trained on large amounts of data, which helps capture world knowledge alongside linguistic competence. Due to this, they are extensively used for ultra-fine entity typing tasks, where they provide the entity knowledge held in its parameter space. Given that PLMs learn from co-occurrence patterns, they likely contain more knowledge or less knowledge about entities depending on their how frequent they are in the pre-training data. In this work, we probe PLMs to elicit encoded entity probabilities and demonstrate that they highly correlate with their frequency in large-scale internet data. Then, we demonstrate that entity-typing approaches that rely on PLMs struggle with entities at the long tail on the distribution. Our findings suggests that we need to go beyond PLMs to produce solutions that perform well for rare, new or infrequent entities.

In recent years, there has been increasing interest in using formal methods-based techniques to safely achieve temporal tasks, such as timed sequence of goals, or patrolling objectives. Such tasks are often expressed in real-time logics such as Signal Temporal Logic (STL), whereby, the logical specification is encoded into an optimization problem. Such approaches usually involve optimizing over the quantitative semantics, or robustness degree, of the logic over bounded horizons: the semantics can be encoded as mixed-integer linear constraints or into smooth approximations of the robustness degree. A major limitation of this approach is that it faces scalability challenges with respect to temporal complexity: for example, encoding long-term tasks requires storing the entire history of the system. In this paper, we present a quantitative generalization of such tasks in the form of symbolic automata objectives. Specifically, we show that symbolic automata can be expressed as matrix operators that lend themselves to automatic differentiation, allowing for the use of off-the-shelf gradient-based optimizers. We show how this helps solve the need to store arbitrarily long system trajectories, while efficiently leveraging the task structure encoded in the automaton., Comment: The paper has been accepted to IROS 2024

The demand for sophisticated natural language processing (NLP) methods, particularly Named Entity Recognition (NER), has increased due to the exponential growth of Marathi-language digital content. In particular, NER is essential for recognizing distant entities and for arranging and understanding unstructured Marathi text data. With an emphasis on managing long-range entities, this paper offers a comprehensive analysis of current NER techniques designed for Marathi documents. It dives into current practices and investigates the BERT transformer model's potential for long-range Marathi NER. Along with analyzing the effectiveness of earlier methods, the report draws comparisons between NER in English literature and suggests adaptation strategies for Marathi literature. The paper discusses the difficulties caused by Marathi's particular linguistic traits and contextual subtleties while acknowledging NER's critical role in NLP. To conclude, this project is a major step forward in improving Marathi NER techniques, with potential wider applications across a range of NLP tasks and domains.

We present the MahaSUM dataset, a large-scale collection of diverse news articles in Marathi, designed to facilitate the training and evaluation of models for abstractive summarization tasks in Indic languages. The dataset, containing 25k samples, was created by scraping articles from a wide range of online news sources and manually verifying the abstract summaries. Additionally, we train an IndicBART model, a variant of the BART model tailored for Indic languages, using the MahaSUM dataset. We evaluate the performance of our trained models on the task of abstractive summarization and demonstrate their effectiveness in producing high-quality summaries in Marathi. Our work contributes to the advancement of natural language processing research in Indic languages and provides a valuable resource for future research in this area using state-of-the-art models. The dataset and models are shared publicly at https://github.com/l3cube-pune/MarathiNLP

Posterior sampling in contextual bandits with a Gaussian prior can be implemented exactly or approximately using the Laplace approximation. The Gaussian prior is computationally efficient but it cannot describe complex distributions. In this work, we propose approximate posterior sampling algorithms for contextual bandits with a diffusion model prior. The key idea is to sample from a chain of approximate conditional posteriors, one for each stage of the reverse process, which are estimated in a closed form using the Laplace approximation. Our approximations are motivated by posterior sampling with a Gaussian prior, and inherit its simplicity and efficiency. They are asymptotically consistent and perform well empirically on a variety of contextual bandit problems., Comment: Proceedings of the 38th Conference on Neural Information Processing Systems

Given a smooth scheme X with an action by an affine algebraic group G, we give a formula to compute the Nisnevich sheaf of the motivic connected components of the quotient stack [X/G] in the case of an orbifold. We apply it to identify all the $\mathbb{A}^1$-connected stacky curves and prove the $\mathbb{A}^1$ -connectedness of the moduli stack of elliptic curves. We also prove homotopy purity for the smooth algebraic stacks and recover a computation of the Picard group of stacky orbifold curves by computing their mixed motive., Comment: Corrected Theorem 1.1, streamlined section 3 and some other minor changes. Comments still welcome

Large language models (LLMs) have revolutionized how we interact with technology, but their personalization to individual user preferences remains a significant challenge, particularly in on-device applications. Traditional methods often depend heavily on labeled datasets and can be resource-intensive. To address these issues, we present Adaptive Self-Supervised Learning Strategies (ASLS), which utilizes self-supervised learning techniques to personalize LLMs dynamically. The framework comprises a user profiling layer for collecting interaction data and a neural adaptation layer for real-time model fine-tuning. This innovative approach enables continuous learning from user feedback, allowing the model to generate responses that align closely with user-specific contexts. The adaptive mechanisms of ASLS minimize computational demands and enhance personalization efficiency. Experimental results across various user scenarios illustrate the superior performance of ASLS in boosting user engagement and satisfaction, highlighting its potential to redefine LLMs as highly responsive and context-aware systems on-device., Comment: First ASLS

Near-sun sky twilight observations allow for the detection of asteroid interior to the orbit of Venus (Aylos), the Earth (Atiras), and comets. We present the results of observations with the Palomar 48-inch telescope (P48)/Zwicky Transient Facility (ZTF) camera in 30 s r-band exposures taken during evening astronomical twilight from 2019 Sep 20 to 2022 March 7 and during morning astronomical twilight sky from 2019 Sep 21 to 2022 Sep 29. More than 46,000 exposures were taken in evening and morning astronomical twilight within 31 to 66 degrees from the Sun with an r-band limiting magnitude between 18.1 and 20.9. The twilight pointings show a slight seasonal dependence in limiting magnitude and ability to point closer towards the Sun, with limiting magnitude slightly improving during summer. In total, the one Aylo, (594913) 'Ayl\'o'chaxnim, and 4 Atiras, 2020 OV1, 2021 BS1, 2021 PB2, and 2021 VR3, were discovered in evening and morning twilight observations. Additional twilight survey discoveries also include 6 long-period comets: C/2020 T2, C/2020 V2, C/2021 D2, C/2021 E3, C/2022 E3, and C/2022 P3, and two short-period comets: P/2021 N1 and P/2022 P2 using deep learning comet detection pipelines. The P48/ZTF twilight survey also recovered 11 known Atiras, one Aylo, three short-period comes, two long-period comets, and one interstellar object. Lastly, the Vera Rubin Observatory will conduct a twilight survey starting in its first year of operations and will cover the sky within 45 degrees of the Sun. Twilight surveys such as those by ZTF and future surveys will provide opportunities for discovering asteroids inside the orbits of Earth and Venus., Comment: 26 pages, 13 figures, 4 tables, accepted for publication in Icarus

Wolf-Rayet stars (WRs) are one of the final evolutionary stages of massive stars and immediate progenitors of stellar-mass black holes. Their multiplicity forms an important anchor point in single and binary population models for predicting gravitational-wave progenitors. Recent spectroscopic campaigns have suggested incompatible multiplicity fractions and period distributions for N- and C-rich Galactic WRs (WNs and WCs) at short as well as long orbital periods, in contradiction with evolutionary model predictions. In this work, we employed infrared interferometry using the $K$-band instrument GRAVITY at the VLTI to investigate the multiplicity of WRs at long periods and explore the nature of their companions. We present a survey of 39 Galactic WRs, including 11 WN, 15 WC and 13 H-rich WN (WNh) stars. We detected wide companions with GRAVITY for only four stars: WR 48, WR 89, WR 93 and WR 115. Combining with spectroscopic studies, we arrived at multiplicity fractions of $f^{\rm WN}_{\rm obs} = 0.55\pm0.15$, $f^{\rm WC}_{\rm obs} = 0.40\pm0.13$ and $f^{\rm WNh}_{\rm obs} = 0.23\pm0.12$. In addition, we also found other features in the GRAVITY dataset such as (i) a diffuse extended component in over half the WR sample; (ii) five known spectroscopic binaries resolved in differential phase data and (iii) spatially resolved winds in four stars: WR 16, WR 31a, WR 78 and WR 110. Our survey reveals a lack of intermediate (few 100s d) and long- (few years to decades) period WR systems. The 200-d peak in the period distributions of WR+OB and BH+OB binaries predicted by Case B mass-transfer binary evolution models is not seen in our data. The rich companionship of their O-type progenitors in this separation range suggest that the WR progenitor stars expand and interact with their companions, most likely through unstable mass-transfer, resulting in either a short-period system or a merger., Comment: Accepted in A&A

We present an analysis of Hubble Space Telescope (HST) data from WFC3/UVIS, WFC3/IR and ACS, investigating the young stellar cluster (YSC) population in the face-on spiral galaxy M83. Within the field of view of the IR pointings, we identify 454 sources with compact F814W continuum and Pa$\beta$ line emission with a S/N $\geq 3$ as possible YSC candidates embedded in dust. We refine this selection to 97 candidates based on their spectral energy distributions, multi-wavelength morphology, and photometric uncertainties. For sources that are detected in all bands and have mass $> 10^{2.8} M_{\odot}$ (53 sources), we find that by 2 Myr $75\%$ of infrared-selected star clusters have an $A_{V} \leq 1$, and that by 3 Myr the fraction rises to $\sim 82\%$. This evidence of early clearing implies that pre-supernovae feedback from massive stars are responsible for clearing the majority of the natal gas and dust that surround infrared-selected star clusters in M83. Further, this result is consistent with previous estimates based on WFC3 observations, and adds to the growing body of literature suggesting pre-supernova feedback to be crucial for YSC emergence in normal star-forming galaxies. Finally, we find a weak correlation between the YSC concentration index and age over the first 10 Myr, which matches previous studies and indicates little or no change in the size of YSCs in M83 during their early evolution., Comment: 14 pages, 5 figures, 2 tables. Accepted for publication in ApJL

In this survey, we design formal verification and control algorithms for autonomous systems with practical safety guarantees using conformal prediction (CP), a statistical tool for uncertainty quantification. We focus on learning-enabled autonomous systems (LEASs) in which the complexity of learning-enabled components (LECs) is a major bottleneck that hampers the use of existing model-based verification and design techniques. Instead, we advocate for the use of CP, and we will demonstrate its use in formal verification, systems and control theory, and robotics. We argue that CP is specifically useful due to its simplicity (easy to understand, use, and modify), generality (requires no assumptions on learned models and data distributions, i.e., is distribution-free), and efficiency (real-time capable and accurate). We pursue the following goals with this survey. First, we provide an accessible introduction to CP for non-experts who are interested in using CP to solve problems in autonomy. Second, we show how to use CP for the verification of LECs, e.g., for verifying input-output properties of neural networks. Third and fourth, we review recent articles that use CP for safe control design as well as offline and online verification of LEASs. We summarize their ideas in a unifying framework that can deal with the complexity of LEASs in a computationally efficient manner. In our exposition, we consider simple system specifications, e.g., robot navigation tasks, as well as complex specifications formulated in temporal logic formalisms. Throughout our survey, we compare to other statistical techniques (e.g., scenario optimization, PAC-Bayes theory, etc.) and how these techniques have been used in verification and control. Lastly, we point the reader to open problems and future research directions.

In this article, we give a unified proof of the end-point estimates of the totally-geodesic $k$-plane transform of radial functions on spaces of constant curvature. The problem of getting end-point estimates is not new and some results are available in literature. However, these results were obtained independently without much focus on the similarities between underlying geometries. We give a unified proof for the end-point estimates on spaces of constant curvature by making use of geometric ideas common to the spaces of constant curvature, and obtaining a unified formula for the $k$-plane transform of radial functions. We also give some inequalities for certain special functions as an application to one of our lemmata., Comment: 15 pages, 4 figures

Waterbomb style tessellations have been explored in the past by artists such as Ronald D. Resch, Benjamin Parker and Mitya Miller. Generalised waterbomb tessellations are still underexplored in origami design. We have explored various sets of criteria for generalising waterbomb tessellations in order to enumerate valid patterns. We only consider triangular waterbomb tessellations, other polygons will be explored in future papers. In our search we have uncovered some new waterbomb tessellations, which could offer new uses in representational and geometric origami design. We conclude by discussing foldability properties and possible generalisations.

With the increasing demand for data sharing across platforms and organizations, ensuring the privacy and security of sensitive information has become a critical challenge. This paper introduces "TableGuard". An innovative approach to data obfuscation tailored for relational databases. Building on the principles and techniques developed in prior work on context-sensitive obfuscation, TableGuard applies these methods to ensure that API calls return only obfuscated data, thereby safeguarding privacy when sharing data with third parties. TableGuard leverages advanced context-sensitive obfuscation techniques to replace sensitive data elements with contextually appropriate alternatives. By maintaining the relational integrity and coherence of the data, our approach mitigates the risks of cognitive dissonance and data leakage. We demonstrate the implementation of TableGuard using a BERT based transformer model, which identifies and obfuscates sensitive entities within relational tables. Our evaluation shows that TableGuard effectively balances privacy protection with data utility, minimizing information loss while ensuring that the obfuscated data remains functionally useful for downstream applications. The results highlight the importance of domain-specific obfuscation strategies and the role of context length in preserving data integrity. The implications of this research are significant for organizations that need to share data securely with external parties. TableGuard offers a robust framework for implementing privacy-preserving data sharing mechanisms, thereby contributing to the broader field of data privacy and security., Comment: 7 pages, 3 tables, 1 figure

In image processing, one of the most challenging tasks is to render an image's semantic meaning using a variety of artistic approaches. Existing techniques for arbitrary style transfer (AST) frequently experience mode-collapse, over-stylization, or under-stylization due to a disparity between the style and content images. We propose a novel framework called D$^2$Styler (Discrete Diffusion Styler) that leverages the discrete representational capability of VQ-GANs and the advantages of discrete diffusion, including stable training and avoidance of mode collapse. Our method uses Adaptive Instance Normalization (AdaIN) features as a context guide for the reverse diffusion process. This makes it easy to move features from the style image to the content image without bias. The proposed method substantially enhances the visual quality of style-transferred images, allowing the combination of content and style in a visually appealing manner. We take style images from the WikiArt dataset and content images from the COCO dataset. Experimental results demonstrate that D$^2$Styler produces high-quality style-transferred images and outperforms twelve existing methods on nearly all the metrics. The qualitative results and ablation studies provide further insights into the efficacy of our technique. The code is available at https://github.com/Onkarsus13/D2Styler., Comment: Paper accepted at 27th International Conference on Pattern Recognition (ICPR), 2024