arXiv subjects

Sandeep Kumar

Publications and source records attributed to Sandeep Kumar.

At least 37 records · Page 2Linked to original sources

arXiv2026

Quantum walk inspired JPEG compression of images

This work proposes a quantum inspired adaptive quantization framework that enhances the classical JPEG compression by introducing a learned, optimized Qtable derived using a Quantum Walk Inspired Optimization (QWIO) search strategy. The optimizer searches a continuous parameter space of frequency band scaling factors under a unified rate distortion objective that jointly considers reconstruction fidelity and compression efficiency. The proposed framework is evaluated on MNIST, CIFAR10, and ImageNet subsets, using Peak Signal to Noise Ratio (PSNR), Structural Similarity Index (SSIM), Bits Per Pixel (BPP), and error heatmap visual analysis as evaluation metrics. Experimental results show average gains ranging from 3 to 6 dB PSNR, along with better structural preservation of edges, contours, and luminance transitions, without modifying decoder compatibility. The structure remains JPEG compliant and can be implemented using accessible scientific packages making it ideal for deployment and practical research use.

eess.IV↗

arXiv2026

Interlayer charge transfer from contact electrification in conducting micro and nanoscale thin film heterostructures

Contact electrification give rise to charge accumulation at the interface when two materials are brought into contact with each other. The charge accumulation at the interface will diffuse to the interior of the conducting material if the dimensions of the contacting conducting material is of the order of an unknown critical length scale. This contact electrification induced interlayer charge transfer will modify the fundamental physical properties of both the contacting materials. This review first discusses the reported experimental evidence of flexoelectricity induced contact electrification and interlayer charge transfer in conducting thin film based heterostructures. The interlayer charge transfer creates a gradient of charge carrier in both the thin films constituting the heterostructure and also modifies the electron-electron interactions. Further, the interlayer charge transfer changes the electron-phonon coupling, spin-phonon coupling and magnetoelectronic coupling that give rise to new physical behavior, which did not exist prior to the interlayer charge transfer. The new physical behaviors from interlayer charge transfer and their mechanistic origins are reanalyzed and discussed, which include spin-Hall effect of charge carriers, topological Hall effect of magnetoelectronic electromagnon, inhomogeneous magnetoelectronic multiferroic effect, flexoelectronic proximity effect and topological spin texture. This review article presents a unified picture of current status and future directions that will provide the scientists a stepping stone for research in the field of flexoelectricity mediated contact electrification and interlayer charge transfer mediated behavior in the micro/nanoscale heterostructures of the conducting materials.

cond-mat.mes-hall↗

arXiv2026

ConstMig: Enabling Secure Live Migration of Large Intel SGX-based applications

Cloud service providers are adopting Trusted Execution Environments (TEEs) to provide hardware-guaranteed security to applications running on remote, untrusted data centers. However, migrating such applications still relies on the decade-old stop-and-copy method, which introduces large downtimes. Modern live-migration approaches such as pre-copy and post-copy do not work for TEE-based applications due to hardware-enforced restrictions. We propose ConstMig, a near-zero-downtime live-migration mechanism for large memory-footprint TEE-based applications. ConstMig is fully compatible with containers, virtual machines (VMs), and microVMs. Our prototype, built on Intel SGX, achieves near-zero downtime irrespective of enclave size and requires no additional hardware support. ConstMig reduces total downtime by 77 - 96% for a suite of SGX applications with multi-gigabyte memory footprints compared to state-of-the-art TEE-based migration solutions such as MigSGX.

cs.CR↗

arXiv2026

GRAPHGINI: Fostering Individual and Group Fairness in Graph Neural Networks

Graph Neural Networks (GNNs) have demonstrated impressive performance across various tasks, leading to their increased adoption in high-stakes decision-making systems. However, concerns have arisen about GNNs potentially generating unfair decisions for underprivileged groups or individuals when lacking fairness constraints. This work addresses this issue by introducing GraphGini, a novel approach that incorporates the Gini coefficient to enhance both individual and group fairness within the GNN framework. We rigorously establish that the Gini coefficient offers greater robustness and promotes equal opportunity among GNN outcomes, advantages not afforded by the prevailing Lipschitz constant methodology. Additionally, we employ the Nash social welfare program to ensure our solution yields a Pareto optimal distribution of group fairness. Extensive experimentation on real-world datasets demonstrates GraphGini's efficacy in significantly improving individual fairness compared to state-of-the-art methods while maintaining utility and group fairness.

cs.LG↗

arXiv2025

AdiBhashaa: A Community-Curated Benchmark for Machine Translation into Indian Tribal Languages

Large language models and multilingual machine translation (MT) systems increasingly drive access to information, yet many languages of the tribal communities remain effectively invisible in these technologies. This invisibility exacerbates existing structural inequities in education, governance, and digital participation. We present AdiBhashaa, a community-driven initiative that constructs the first open parallel corpora and baseline MT systems for four major Indian tribal languages-Bhili, Mundari, Gondi, and Santali. This work combines participatory data creation with native speakers, human-in-the-loop validation, and systematic evaluation of both encoder-decoder MT models and large language models. In addition to reporting technical findings, we articulate how AdiBhashaa illustrates a possible model for more equitable AI research: it centers local expertise, builds capacity among early-career researchers from marginalized communities, and foregrounds human validation in the development of language technologies.

cs.CL↗

arXiv2025

CrossMed: A Multimodal Cross-Task Benchmark for Compositional Generalization in Medical Imaging

Recent advances in multimodal large language models have enabled unified processing of visual and textual inputs, offering promising applications in general-purpose medical AI. However, their ability to generalize compositionally across unseen combinations of imaging modality, anatomy, and task type remains underexplored. We introduce CrossMed, a benchmark designed to evaluate compositional generalization (CG) in medical multimodal LLMs using a structured Modality-Anatomy-Task (MAT) schema. CrossMed reformulates four public datasets, CheXpert (X-ray classification), SIIM-ACR (X-ray segmentation), BraTS 2020 (MRI classification and segmentation), and MosMedData (CT classification) into a unified visual question answering (VQA) format, resulting in 20,200 multiple-choice QA instances. We evaluate two open-source multimodal LLMs, LLaVA-Vicuna-7B and Qwen2-VL-7B, on both Related and Unrelated MAT splits, as well as a zero-overlap setting where test triplets share no Modality, Anatomy, or Task with the training data. Models trained on Related splits achieve 83.2 percent classification accuracy and 0.75 segmentation cIoU, while performance drops significantly under Unrelated and zero-overlap conditions, demonstrating the benchmark difficulty. We also show cross-task transfer, where segmentation performance improves by 7 percent cIoU even when trained using classification-only data. Traditional models (ResNet-50 and U-Net) show modest gains, confirming the broad utility of the MAT framework, while multimodal LLMs uniquely excel at compositional generalization. CrossMed provides a rigorous testbed for evaluating zero-shot, cross-task, and modality-agnostic generalization in medical vision-language models.

cs.CV↗

arXiv2025

Enhancing Robustness of Graph Neural Networks through p-Laplacian

With the increase of data in day-to-day life, businesses and different stakeholders need to analyze the data for better predictions. Traditionally, relational data has been a source of various insights, but with the increase in computational power and the need to understand deeper relationships between entities, the need to design new techniques has arisen. For this graph data analysis has become an extraordinary tool for understanding the data, which reveals more realistic and flexible modelling of complex relationships. Recently, Graph Neural Networks (GNNs) have shown great promise in various applications, such as social network analysis, recommendation systems, drug discovery, and more. However, many adversarial attacks can happen over the data, whether during training (poisoning attack) or during testing (evasion attack), which can adversely manipulate the desired outcome from the GNN model. Therefore, it is crucial to make the GNNs robust to such attacks. The existing robustness methods are computationally demanding and perform poorly when the intensity of attack increases. This paper presents a computationally efficient framework, namely, pLAPGNN, based on weighted p-Laplacian for making GNNs robust. Empirical evaluation on real datasets establishes the efficacy and efficiency of the proposed method.

cs.LG↗

arXiv2025

Leveraging the Cross-Domain & Cross-Linguistic Corpus for Low Resource NMT: A Case Study On Bhili-Hindi-English Parallel Corpus

The linguistic diversity of India poses significant machine translation challenges, especially for underrepresented tribal languages like Bhili, which lack high-quality linguistic resources. This paper addresses the gap by introducing Bhili-Hindi-English Parallel Corpus (BHEPC), the first and largest parallel corpus worldwide comprising 110,000 meticulously curated sentences across Bhili, Hindi, and English. The corpus was created with the assistance of expert human translators. BHEPC spans critical domains such as education, administration, and news, establishing a valuable benchmark for research in low resource machine translation. To establish a comprehensive Bhili Machine Translation benchmark, we evaluated a wide range of proprietary and open-source Multilingual Large Language Models (MLLMs) on bidirectional translation tasks between English/Hindi and Bhili. Comprehensive evaluation demonstrates that the fine-tuned NLLB-200 distilled 600M variant model outperforms others, highlighting the potential of multilingual models in low resource scenarios. Furthermore, we investigated the generative translation capabilities of multilingual LLMs on BHEPC using in-context learning, assessing performance under cross-domain generalization and quantifying distributional divergence. This work bridges a critical resource gap and promotes inclusive natural language processing technologies for low-resource and marginalized languages globally.

cs.CL↗

arXiv2025

Can Large Language Models Unlock Novel Scientific Research Ideas?

The widespread adoption of Large Language Models (LLMs) and publicly available ChatGPT have marked a significant turning point in the integration of Artificial Intelligence (AI) into people's everyday lives. This study examines the ability of Large Language Models (LLMs) to generate future research ideas from scientific papers. Unlike tasks such as summarization or translation, idea generation lacks a clearly defined reference set or structure, making manual evaluation the default standard. However, human evaluation in this setting is extremely challenging ie: it requires substantial domain expertise, contextual understanding of the paper, and awareness of the current research landscape. This makes it time-consuming, costly, and fundamentally non-scalable, particularly as new LLMs are being released at a rapid pace. Currently, there is no automated evaluation metric specifically designed for this task. To address this gap, we propose two automated evaluation metrics: Idea Alignment Score (IAScore) and Idea Distinctness Index. We further conducted human evaluation to assess the novelty, relevance, and feasibility of the generated future research ideas. This investigation offers insights into the evolving role of LLMs in idea generation, highlighting both its capability and limitations. Our work contributes to the ongoing efforts in evaluating and utilizing language models for generating future research ideas. We make our datasets and codes publicly available

cs.CL↗

arXiv2025

Histogram Driven Amplitude Embedding for Qubit Efficient Quantum Image Compression

This work introduces a compact and hardware efficient method for compressing color images using near term quantum devices. The approach segments the image into fixed size blocks called bixels, and computes the total intensity within each block. A global histogram with B bins is then constructed from these block intensities, and the normalized square roots of the bin counts are encoded as amplitudes into an n qubit quantum state. Amplitude embedding is performed using PennyLane and executed on real IBM Quantum hardware. The resulting state is measured to reconstruct the histogram, enabling approximate recovery of block intensities and full image reassembly. The method maintains a constant qubit requirement based solely on the number of histogram bins, independent of the resolution of the image. By adjusting B, users can control the trade off between fidelity and resource usage. Empirical results demonstrate high quality reconstructions using as few as 5 to 7 qubits, significantly outperforming conventional pixel level encodings in terms of qubit efficiency and validating the practical application of the method for current NISQ era quantum systems.

quant-ph↗

arXiv2025

Quantum Fourier Transform Based Denoising: Unitary Filtering for Enhanced Speech Clarity

This paper introduces a quantum-inspired denoising framework that integrates the Quantum Fourier Transform (QFT) into classical audio enhancement pipelines. Unlike conventional Fast Fourier Transform (FFT) based methods, QFT provides a unitary transformation with global phase coherence and energy preservation, enabling improved discrimination between speech and noise. The proposed approach replaces FFT in Wiener and spectral subtraction filters with a QFT operator, ensuring consistent hyperparameter settings for fair comparison. Experiments on clean speech, synthetic tones, and noisy mixtures across diverse signal to noise ratio (SNR) conditions, demonstrate statistically significant gains in SNR, with up to 15 dB improvement and reduced artifact generation. Results confirm that QFT based denoising offers robustness under low SNR and nonstationary noise scenarios without additional computational overhead, highlighting its potential as a scalable pathway toward quantum-enhanced speech processing.

cs.SD↗

arXiv2025

Transforming Blood Cell Detection and Classification with Advanced Deep Learning Models: A Comparative Study

Efficient detection and classification of blood cells are vital for accurate diagnosis and effective treatment of blood disorders. This study utilizes a YOLOv10 model trained on Roboflow data with images resized to 640x640 pixels across varying epochs. The results show that increased training epochs significantly enhance accuracy, precision, and recall, particularly in real-time blood cell detection & classification. The YOLOv10 model outperforms MobileNetV2, ShuffleNetV2, and DarkNet in real-time performance, though MobileNetV2 and ShuffleNetV2 are more computationally efficient, and DarkNet excels in feature extraction for blood cell classification. This research highlights the potential of integrating deep learning models like YOLOv10, MobileNetV2, ShuffleNetV2, and DarkNet into clinical workflows, promising improvements in diagnostic accuracy and efficiency. Additionally, a new, well-annotated blood cell dataset was created and will be open-sourced to support further advancements in automatic blood cell detection and classification. The findings demonstrate the transformative impact of these models in revolutionizing medical diagnostics and enhancing blood disorder management

eess.IV↗

arXiv2025

Dynamically phase-separated states in driven binary dusty plasma

We comprehensively study external forcing-driven dynamical structure formation in a binary dusty plasma mixture. Using two-dimensional driven-dissipative molecular dynamics simulations, we demonstrate phase segregation into bands and lanes beyond a critical forcing threshold. The particles interact via the Debye-Hückel potential, with interaction strength serving as a control parameter for determining the critical forcing. During early evolution, the results exhibit features of two-stream instability. A steady-state phase-space diagram indicates that bands and lanes emerge beyond a critical forcing and coupling strength. Lanes predominantly form under high external forcing. Multiple independent diagnostics, including the order parameter, drift velocity, diffusion coefficients, domain size, and the final-to-initial coupling strength ratio, provide insight into phase segregation and help determine the critical forcing amplitude. Furthermore, we show that the time evolution of band and lane widths follows an exponent of 1/3 for both critical and off-critical mixtures. These findings contrast with the previously reported scaling of 1/2 for equilibrium phase separation in critical mixtures. These results help bridge the gap between dusty plasmas and colloidal systems and facilitate controlled dusty plasma experiments in this direction.

physics.plasm-ph↗

arXiv2025

Inflated hot Jupiters: Inferring average atmospheric velocity via Ohmic models coupled with internal dynamo evolution

The inflated radii observed in hundreds of hot Jupiters (HJ) represent a long-standing open issue. In this study, we quantitatively investigate this phenomenon within the framework of Ohmic dissipation arising from magnetic induction in the atmosphere, one of the most promising mechanisms for explaining the radius anomaly. We simulate the evolution of irradiated giant planets with MESA, spanning the observed range of masses and equilibrium temperatures, incorporating an internal source of Ohmic dissipation that extends to deep layers of the envelope. We infer average atmospheric wind intensities, averaged in the region $p < 10$ bar, in the range 0.01-1 km/s in order to reproduce the range of observed radii, decreasing roughly linearly with planetary mass, and much more steeply with equilibrium temperature. This is consistent with the expected effects of magnetic drag from the induced field, which is higher for more intense irradiation, via conductivity, and for larger masses, which have higher dynamo fields. Due to the evolution of the dynamo field and the proportionality of the induced currents on it, the Ohmic efficiency typically decreases by at least one order of magnitude from 0.1 to 10 Gyr, at contrast with the common assumption of a constant-in-time value. Notably, the extent of the main convective region, and the associated heat flux supporting the dynamo, is reduced in the presence of strong Ohmic dissipation, which in turn depends on the dynamo field strength, generating a non-trivial coupling of the latter with the atmospheric induction, potentially leading to an oscillatory behaviour of the field strength. These findings remain generally valid even when accounting for a long-term increase in the main-sequence host star luminosity, although this case can more readily lead to HJ re-inflation, consistent with previous studies.

astro-ph.EP↗

arXiv2025

Graph-Coarsening for Machine Learning Coarse-grained Molecular Dynamics

Coarse-grained (CG) molecular dynamics (MD) simulations can simulate large molecular complexes over extended timescales by reducing degrees of freedom. A critical step in CG modeling is the selection of the CG mapping algorithm, which directly influences both accuracy and interpretability of the model. Despite progress, the optimal strategy for coarse-graining remains a challenging task, highlighting the necessity for a comprehensive theoretical framework. In this work, we present a graph-based coarsening approach to develop CG models. Coarse-grained sites are obtained through edge contractions, where nodes are merged based on a local variational cost metric while preserving key spectral properties of the original graph. Furthermore, we illustrate how Message Passing Atomic Cluster Expansion (MACE) can be applied to generate ML-CG potentials that are not only highly efficient but also accurate. Our approach provides a bottom-up, theoretically grounded computational method for the development of systematically improvable CG potentials.

cond-mat.soft↗

arXiv2025

A Hybrid Ensemble Learning Framework for Image-Based Solar Panel Classification

The installation of solar energy systems is on the rise, and therefore, appropriate maintenance techniques are required to be used in order to maintain maximum performance levels. One of the major challenges is the automated discrimination between clean and dirty solar panels. This paper presents a novel Dual Ensemble Neural Network (DENN) to classify solar panels using image-based features. The suggested approach utilizes the advantages offered by various ensemble models by integrating them into a dual framework, aimed at improving both classification accuracy and robustness. The DENN model is evaluated in comparison to current ensemble methods, showcasing its superior performance across a range of assessment metrics. The proposed approach performs the best compared to other methods and reaches state-of-the-art accuracy on experimental results for the Deep Solar Eye dataset, effectively serving predictive maintenance purposes in solar energy systems. It reveals the potential of hybrid ensemble learning techniques to further advance the prospects of automated solar panel inspections as a scalable solution to real-world challenges.

cs.IT↗

arXiv2025

Quantum-Enhanced Hybrid Reinforcement Learning Framework for Dynamic Path Planning in Autonomous Systems

In this paper, a novel quantum classical hybrid framework is proposed that synergizes quantum with Classical Reinforcement Learning. By leveraging the inherent parallelism of quantum computing, the proposed approach generates robust Q tables and specialized turn cost estimations, which are then integrated with a classical Reinforcement Learning pipeline. The Classical Quantum fusion results in rapid convergence of training, reducing the training time significantly and improved adaptability in scenarios featuring static, dynamic, and moving obstacles. Simulator based evaluations demonstrate significant enhancements in path efficiency, trajectory smoothness, and mission success rates, underscoring the potential of framework for real time, autonomous navigation in complex and unpredictable environments. Furthermore, the proposed framework was tested beyond simulations on practical scenarios, including real world map data such as the IIT Delhi campus, reinforcing its potential for real time, autonomous navigation in complex and unpredictable environments.

cs.LG↗

arXiv2025

A Hybrid Quantum Classical Pipeline for X Ray Based Fracture Diagnosis

Bone fractures are a leading cause of morbidity and disability worldwide, imposing significant clinical and economic burdens on healthcare systems. Traditional X ray interpretation is time consuming and error prone, while existing machine learning and deep learning solutions often demand extensive feature engineering, large, annotated datasets, and high computational resources. To address these challenges, a distributed hybrid quantum classical pipeline is proposed that first applies Principal Component Analysis (PCA) for dimensionality reduction and then leverages a 4 qubit quantum amplitude encoding circuit for feature enrichment. By fusing eight PCA derived features with eight quantum enhanced features into a 16 dimensional vector and then classifying with different machine learning models achieving 99% accuracy using a public multi region X ray dataset on par with state of the art transfer learning models while reducing feature extraction time by 82%.

eess.IV↗