arXiv subjects

Vishal Sharma

Publications and source records attributed to Vishal Sharma.

At least 19 recordsLinked to original sources

arXiv2026

Physics-Informed Neural Networks for Depth-Averaged Granular Avalanche Dynamics on Curved Topography

Physics-informed neural networks (PINNs) provide a mesh-free framework for solving governing equations, but their application to granular avalanche dynamics over curved terrain remains largely unexplored. This study extends a depth-averaged PINN formulation based on the Savage-Hutter equations to an exponentially curved chute with spatially varying inclination and a strain-rate-dependent Mohr-Coulomb earth-pressure closure. The model is validated against measured front- and rear-edge trajectories from a laboratory granular-avalanche experiment, with selected observations withheld from training. A staged temporal curriculum proved essential for accurate prediction, reducing the held-out trajectory error by approximately two orders of magnitude compared with training over the full time domain from the outset. Sparse-data experiments further showed that observation placement was more influential than observation number within the configurations tested. Four observations bracketing the transition from acceleration to deceleration achieved nearly the same accuracy as the eight-observation reference configuration, whereas observations clustered at early or late times performed poorly. The results demonstrate the importance of both training strategy and informative data placement when applying PINNs to granular flows over curved topography.

cond-mat.soft↗

arXiv2026

PULSAR: Pooled Unified Late-Interaction Search and Retrieval for Enterprise Visual Document RAG

Institutional investors search visually dense pitch decks, board packs, and diligence materials that change hourly near deal closing. OCR followed by figure verbalisation is costly to refresh at this scale and can lose chart detail. We present PULSAR, a production vision-first retrieval system deployed at Mubadala Investment Company. PULSAR indexes page images with a frozen ColPali-style backbone and uses a pooled two-stage late-interaction index: compact page summaries support initial retrieval, followed by exact MaxSim rescoring over a finer pooled representation. On ViDoRe V3, this design reduces median vector-search latency by 15.1 times against an unpooled configuration with less than 0.01 absolute NDCG@10 and Recall@10 loss; production median vector-search latency is 156 ms. Under concurrent load, the pooled index sustains approximately 88 times higher QPS than an unpooled index. The event-driven ingestion path is estimated to be approximately 20 times cheaper per page than the OCR+verbalisation baseline it replaced. Since March 2026, PULSAR has served 78 thousand documents and approximately 2.4 million pages across more than 3,000 deals. At the production top K, it more than doubles answer-fact recall over the OCR+verbalisation baseline.

cs.IR↗

arXiv2026

Improving Certified Robustness via Adversarial Distillation

Certified training aims to produce models whose predictions can be formally verified against adversarial perturbations, typically by optimising upper bounds on the worst-case loss over an allowed perturbation set. For neural networks, certified training methods based purely on tight relaxation bounds produce networks that are amenable to certification, but sacrifice standard accuracy. Conversely, adversarial training often yields stronger empirical robustness and standard accuracy, but the resulting models are generally difficult to certify with neural network verifiers. Recently, the literature has shown that better standard-certified accuracy trade-offs can be achieved by combining adversarial training objectives with loose over-approximations based on Interval Bound Propagation (IBP), effectively interpolating between lower and upper bounds of the worst-case loss. Building on this, we introduce AD-CERT, a certified training objective that combines adversarial distillation with an IBP upper bound. We show that distilling adversarial information over the logit space from an empirically robust teacher provides an effective lower bound surrogate for certified training, with AD-CERT achieving state-of-the-art certified performance on several robustness benchmarks. Furthermore, in a unified setup, distilling adversarial information at the logit-level is shown to improve certified accuracy over a robust feature-space distillation objective by up to 5.40 percentage points.

cs.LG↗

arXiv2026

UnWEIRDing Peer Review in Human Computer Interaction

Peer review determines which scholarship is legitimized; however, review biases often disadvantage scholarship that diverges from the norm. Human-Computer Interaction (HCI) lacks a systemic inquiry into how such biases affect underrepresented Global South (GS) scholarship. To address this critical gap, we conducted four focus groups with 16 HCI researchers studying the GS. Participants reported experiencing reviews that confined them to development research, dismissed their theoretical contributions, and questioned situated knowledge from GS communities. Both as authors and reviewers, participants reported experiencing the epistemic burden of over-explaining why knowledge from GS communities matters. Further, they noted being tokenized as ``cultural experts'' when assigned to review papers and pointed out that the hidden curriculum of writing HCI papers often gatekeeps GS scholarship. Using epistemic oppression as a lens, we discuss how review practices marginalize GS scholarship and outline actionable strategies for nurturing equitable epistemological evaluation of HCI scholarship.

cs.HC↗

arXiv2026

Performance of a Deep Learning-Based Segmentation Model for Pancreatic Tumors on Public Endoscopic Ultrasound Datasets

Background: Pancreatic cancer is one of the most aggressive cancers, with poor survival rates. Endoscopic ultrasound (EUS) is a key diagnostic modality, but its effectiveness is constrained by operator subjectivity. This study evaluates a Vision Transformer-based deep learning segmentation model for pancreatic tumors. Methods: A segmentation model using the USFM framework with a Vision Transformer backbone was trained and validated with 17,367 EUS images (from two public datasets) in 5-fold cross-validation. The model was tested on an independent dataset of 350 EUS images from another public dataset, manually segmented by radiologists. Preprocessing included grayscale conversion, cropping, and resizing to 512x512 pixels. Metrics included Dice similarity coefficient (DSC), intersection over union (IoU), sensitivity, specificity, and accuracy. Results: In 5-fold cross-validation, the model achieved a mean DSC of 0.651 +/- 0.738, IoU of 0.579 +/- 0.658, sensitivity of 69.8%, specificity of 98.8%, and accuracy of 97.5%. For the external validation set, the model achieved a DSC of 0.657 (95% CI: 0.634-0.769), IoU of 0.614 (95% CI: 0.590-0.689), sensitivity of 71.8%, and specificity of 97.7%. Results were consistent, but 9.7% of cases exhibited erroneous multiple predictions. Conclusions: The Vision Transformer-based model demonstrated strong performance for pancreatic tumor segmentation in EUS images. However, dataset heterogeneity and limited external validation highlight the need for further refinement, standardization, and prospective studies.

cs.CV↗

arXiv2025

Evolving Excellence: Automated Optimization of LLM-based Agents

Agentic AI systems built on large language models (LLMs) offer significant potential for automating complex workflows, from software development to customer support. However, LLM agents often underperform due to suboptimal configurations; poorly tuned prompts, tool descriptions, and parameters that typically require weeks of manual refinement. Existing optimization methods either are too complex for general use or treat components in isolation, missing critical interdependencies. We present ARTEMIS, a no-code evolutionary optimization platform that jointly optimizes agent configurations through semantically-aware genetic operators. Given only a benchmark script and natural language goals, ARTEMIS automatically discovers configurable components, extracts performance signals from execution logs, and evolves configurations without requiring architectural modifications. We evaluate ARTEMIS on four representative agent systems: the \emph{ALE Agent} for competitive programming on AtCoder Heuristic Contest, achieving a \textbf{$13.6\%$ improvement} in acceptance rate; the \emph{Mini-SWE Agent} for code optimization on SWE-Perf, with a statistically significant \textbf{10.1\% performance gain}; and the \emph{CrewAI Agent} for cost and mathematical reasoning on Math Odyssey, achieving a statistically significant \textbf{$36.9\%$ reduction} in the number of tokens required for evaluation. We also evaluate the \emph{MathTales-Teacher Agent} powered by a smaller open-source model (Qwen2.5-7B) on GSM8K primary-level mathematics problems, achieving a \textbf{22\% accuracy improvement} and demonstrating that ARTEMIS can optimize agents based on both commercial and local models.

cs.SE↗

arXiv2025

Self-Adaptive Graph Mixture of Models

Graph Neural Networks (GNNs) have emerged as powerful tools for learning over graph-structured data, yet recent studies have shown that their performance gains are beginning to plateau. In many cases, well-established models such as GCN and GAT, when appropriately tuned, can match or even exceed the performance of more complex, state-of-the-art architectures. This trend highlights a key limitation in the current landscape: the difficulty of selecting the most suitable model for a given graph task or dataset. To address this, we propose Self-Adaptive Graph Mixture of Models (SAGMM), a modular and practical framework that learns to automatically select and combine the most appropriate GNN models from a diverse pool of architectures. Unlike prior mixture-of-experts approaches that rely on variations of a single base model, SAGMM leverages architectural diversity and a topology-aware attention gating mechanism to adaptively assign experts to each node based on the structure of the input graph. To improve efficiency, SAGMM includes a pruning mechanism that reduces the number of active experts during training and inference without compromising performance. We also explore a training-efficient variant in which expert models are pretrained and frozen, and only the gating and task-specific layers are trained. We evaluate SAGMM on 16 benchmark datasets covering node classification, graph classification, regression, and link prediction tasks, and demonstrate that it consistently outperforms or matches leading GNN baselines and prior mixture-based methods, offering a robust and adaptive solution for real-world graph learning.

cs.LG↗

arXiv2025

Adaptive LLM Routing under Budget Constraints

Large Language Models (LLMs) have revolutionized natural language processing, but their varying capabilities and costs pose challenges in practical applications. LLM routing addresses this by dynamically selecting the most suitable LLM for each query/task. Previous approaches treat this as a supervised learning problem, assuming complete knowledge of optimal query-LLM pairings. However, real-world scenarios lack such comprehensive mappings and face evolving user queries. We thus propose to study LLM routing as a contextual bandit problem, enabling adaptive decision-making using bandit feedback without requiring exhaustive inference across all LLMs for all queries (in contrast to supervised routing). To address this problem, we develop a shared embedding space for queries and LLMs, where query and LLM embeddings are aligned to reflect their affinity. This space is initially learned from offline human preference data and refined through online bandit feedback. We instantiate this idea through Preference-prior Informed Linucb fOr adaptive rouTing (PILOT), a novel extension of LinUCB. To handle diverse user budgets for model routing, we introduce an online cost policy modeled as a multi-choice knapsack problem, ensuring resource-efficient routing.

cs.LG↗

arXiv2025

Policy Design in Zero-Trust Distributed Networks: Challenges and Solutions

Traditional security architectures are becoming more vulnerable to distributed attacks due to significant dependence on trust. This will further escalate when implementing agentic AI within the systems, as more components must be secured over a similar distributed space. These scenarios can be observed in consumer technologies, such as the dense Internet of things (IoT). Here, zero-trust architecture (ZTA) can be seen as a potential solution, which relies on a key principle of not giving users explicit trust, instead always verifying their privileges whenever a request is made. However, the overall security in ZTA is managed through its policies, and unverified policies can lead to unauthorized access. Thus, this paper explores challenges and solutions for ZTA policy design in the context of distributed networks, which is referred to as zero-trust distributed networks (ZTDN). This is followed by a case-study on formal verification of policies using UPPAAL. Subsequently, the importance of accountability and responsibility in the system's security is discussed.

cs.NI↗

arXiv2025

RSLAQ -- A Robust SLA-driven 6G O-RAN QoS xApp using deep reinforcement learning

The evolution of 6G envisions a wide range of applications and services characterized by highly differentiated and stringent Quality of Service (QoS) requirements. Open Radio Access Network (O-RAN) technology has emerged as a transformative approach that enables intelligent software-defined management of the RAN. A cornerstone of O-RAN is the RAN Intelligent Controller (RIC), which facilitates the deployment of intelligent applications (xApps and rApps) near the radio unit. In this context, QoS management through O-RAN has been explored using network slice and machine learning (ML) techniques. Although prior studies have demonstrated the ability to optimize RAN resource allocation and prioritize slices effectively, they have not considered the critical integration of Service Level Agreements (SLAs) into the ML learning process. This omission can lead to suboptimal resource utilization and, in many cases, service outages when target Key Performance Indicators (KPIs) are not met. This work introduces RSLAQ, an innovative xApp designed to ensure robust QoS management for RAN slicing while incorporating SLAs directly into its operational framework. RSLAQ translates operator policies into actionable configurations, guiding resource distribution and scheduling for RAN slices. Using deep reinforcement learning (DRL), RSLAQ dynamically monitors RAN performance metrics and computes optimal actions, embedding SLA constraints to mitigate conflicts and prevent outages. Extensive system-level simulations validate the efficacy of the proposed solution, demonstrating its ability to optimize resource allocation, improve SLA adherence, and maintain operational reliability (>95%) in challenging scenarios.

cs.NI↗

arXiv2024

Uncertainty in Supply Chain Digital Twins: A Quantum-Classical Hybrid Approach

This study investigates uncertainty quantification (UQ) using quantum-classical hybrid machine learning (ML) models for applications in complex and dynamic fields, such as attaining resiliency in supply chain digital twins and financial risk assessment. Although quantum feature transformations have been integrated into ML models for complex data tasks, a gap exists in determining their impact on UQ within their hybrid architectures (quantum-classical approach). This work applies existing UQ techniques for different models within a hybrid framework, examining how quantum feature transformation affects uncertainty propagation. Increasing qubits from 4 to 16 shows varied model responsiveness to outlier detection (OD) samples, which is a critical factor for resilient decision-making in dynamic environments. This work shows how quantum computing techniques can transform data features for UQ, particularly when combined with classical methods.

cs.LG↗

arXiv2024

Learning Disentangled Representation in Object-Centric Models for Visual Dynamics Prediction via Transformers

Recent work has shown that object-centric representations can greatly help improve the accuracy of learning dynamics while also bringing interpretability. In this work, we take this idea one step further, ask the following question: "can learning disentangled representation further improve the accuracy of visual dynamics prediction in object-centric models?" While there has been some attempt to learn such disentangled representations for the case of static images \citep{nsb}, to the best of our knowledge, ours is the first work which tries to do this in a general setting for video, without making any specific assumptions about the kind of attributes that an object might have. The key building block of our architecture is the notion of a {\em block}, where several blocks together constitute an object. Each block is represented as a linear combination of a given number of learnable concept vectors, which is iteratively refined during the learning process. The blocks in our model are discovered in an unsupervised manner, by attending over object masks, in a style similar to discovery of slots \citep{slot_attention}, for learning a dense object-centric representation. We employ self-attention via transformers over the discovered blocks to predict the next state resulting in discovery of visual dynamics. We perform a series of experiments on several benchmark 2-D, and 3-D datasets demonstrating that our architecture (1) can discover semantically meaningful blocks (2) help improve accuracy of dynamics prediction compared to SOTA object-centric models (3) perform significantly better in OOD setting where the specific attribute combinations are not seen earlier during training. Our experiments highlight the importance discovery of disentangled representation for visual dynamics prediction.

cs.CV↗

arXiv2024

Fault-tolerant embedding of quantum circuits on hardware architectures via swap gates

In near-term quantum computing devices, connectivity between qubits remain limited by architectural constraints. A computational circuit with given connectivity requirements necessary for multi-qubit gates have to be embedded within physical hardware with fixed connectivity. Long-distance gates have to be done by first routing the relevant qubits together. The simplest routing strategy involves the use of swap gates to swap the information carried by two unconnected qubits to connected ones. Ideal swap gates just permute the qubits; real swap gates, however, have the added possibilities of causing simultaneous errors on the qubits involved and spreading errors across the circuit. A general swap scheme thus changes the error-propagation properties of a circuit, including those necessary for fault-tolerant functioning of a circuit. Here, we present a simple strategy to design the swap scheme needed to embed an abstract circuit onto a physical hardware with constrained connectivity, in a manner that preserves the fault-tolerant properties of the abstract circuit. The embedded circuit will, of course, be noisier, compared to a native implementation of the abstract circuit, but we show in the examples of embedding surface codes on heavy-hexagonal and hexagonal lattices that the deterioration is not severe. This then offers a straightforward solution to implementing circuits with fault-tolerance properties on current hardware.

quant-ph↗

arXiv2024

Sustainable business decision modelling with blockchain and digital twins: A survey

Industry 4.0 and beyond will rely heavily on sustainable Business Decision Modelling (BDM) that can be accelerated by blockchain and Digital Twin (DT) solutions. BDM is built on models and frameworks refined by key identification factors, data analysis, and mathematical or computational aspects applicable to complex business scenarios. Gaining actionable intelligence from collected data for BDM requires a carefully considered infrastructure to ensure data transparency, security, accessibility and sustainability. Organisations should consider social, economic and environmental factors (based on the triple bottom line approach) to ensure sustainability when integrating such an infrastructure. These sustainability features directly impact BDM concerning resource optimisation, stakeholder engagement, regulatory compliance and environmental impacts. To further understand these segments, taxonomies are defined to evaluate blockchain and DT sustainability features based on an in-depth review of the current state-of-the-art research. Detailed comparative evaluations provide insight into the reachability of the sustainable solution in terms of ideologies, access control and performance overheads. Several research questions are put forward to motivate further research that significantly impacts BDM. Finally, a case study based on an exemplary supply chain management system is presented to show the interoperability of blockchain and DT with BDM.

cs.NI↗

arXiv2024

Misconfiguration in O-RAN: Analysis of the impact of AI/ML

User demand on network communication infrastructure has never been greater with applications such as extended reality, holographic telepresence, and wireless brain-computer interfaces challenging current networking capabilities. Open RAN (O-RAN) is critical to supporting new and anticipated uses of 6G and beyond. It promotes openness and standardisation, increased flexibility through the disaggregation of Radio Access Network (RAN) components, supports programmability, flexibility, and scalability with technologies such as Software-Defined Networking (SDN), Network Function Virtualization (NFV), and cloud, and brings automation through the RAN Intelligent Controller (RIC). Furthermore, the use of xApps, rApps, and Artificial Intelligence/Machine Learning (AI/ML) within the RIC enables efficient management of complex RAN operations. However, due to the open nature of O-RAN and its support for heterogeneous systems, the possibility of misconfiguration problems becomes critical. In this paper, we present a thorough analysis of the potential misconfiguration issues in O-RAN with respect to integration and operation, the use of SDN and NFV, and, specifically, the use of AI/ML. The opportunity for AI/ML to be used to identify these misconfigurations is investigated. A case study is presented to illustrate the direct impact on the end user of conflicting policies amongst xApps along with a potential AI/ML-based solution to this problem. This research presents a first analysis of the impact of AI/ML on misconfiguration challenges in O-RAN.

cs.NI↗

arXiv2024

What-if Analysis Framework for Digital Twins in 6G Wireless Network Management

This study explores implementing a digital twin network (DTN) for efficient 6G wireless network management, aligning with the fault, configuration, accounting, performance, and security (FCAPS) model. The DTN architecture comprises the Physical Twin Layer, implemented using NS-3, and the Service Layer, featuring machine learning and reinforcement learning for optimizing carrier sensitivity threshold and transmit power control in wireless networks. We introduce a robust "What-if Analysis" module, utilizing conditional tabular generative adversarial network (CTGAN) for synthetic data generation to mimic various network scenarios. These scenarios assess four network performance metrics: throughput, latency, packet loss, and coverage. Our findings demonstrate the efficiency of the proposed what-if analysis framework in managing complex network conditions, highlighting the importance of the scenario-maker step and the impact of twinning intervals on network performance.

cs.NI↗

arXiv2023

Analysis of Single Photon Detectors in Differential Phase Shift Quantum Key Distribution

In the current research work, an analysis of differential phase shift quantum key distribution using InGaAs/InP and Silicon-APD (avalanche photodiode) as single photon detectors is performed. Various performance parameters of interest such as shifted key rate, secure key rate, and secure communication distance obtained are investigated. In this optical fiber-based differential phase shift quantum key distribution, it is observed that Si-APD under frequency conversion method at telecommunication window outperforms the InGaAs/InP APD.

quant-ph↗

arXiv2022

OpenQAOA -- An SDK for QAOA

We introduce OpenQAOA, a Python open-source multi-backend Software Development Kit to create, customise, and execute the Quantum Approximate Optimisation Algorithm (QAOA) on Noisy Intermediate-Scale Quantum (NISQ) devices and simulators. OpenQAOA facilitates the creation of QAOA workflows, removing the more tedious and repetitive aspects of implementing variational quantum algorithms. It standardises and automates tasks such as circuit creation across different backends, ansatz parametrisation, the optimisation loop, the formatting of results, and extensions of QAOA such as Recursive QAOA. OpenQAOA is designed to simplify and enhance research on QAOA, providing a robust and consistent framework for experimentation with, and deployment of, the algorithm and its variations. Importantly, a heavy emphasis is placed on the provision of tools to enable QAOA computations at the scale of hundreds or thousands of qubits.

quant-ph↗