arXiv subjects

Vivek Sharma

Publications and source records attributed to Vivek Sharma.

At least 19 recordsLinked to original sources

arXiv2025

A Hybrid Ai Framework For Strategic Patent Portfolio Pruning: Integrating Learning To-Rank And Market Need Analysis For Technology Transfer Optimization

This paper introduces a novel, multi stage hybrid intelligence framework for pruning patent portfolios to identify high value assets for technology transfer. Current patent valuation methods often rely on retrospective indicators or manual, time intensive analysis. Our framework automates and deepens this process by combining a Learning to Rank (LTR) model, which evaluates patents against over 30 legal and commercial parameters, with a unique "Need-Seed" agent-based system. The "Need Agent" uses Natural Language Processing (NLP) to mine unstructured market and industry data, identifying explicit technological needs. Concurrently, the "Seed Agent" employs fine tuned Large Language Models (LLMs) to analyze patent claims and map their technological capabilities. The system generates a "Core Ontology Framework" that matches high potential patents (Seeds) to documented market demands (Needs), providing a strategic rationale for divestment decisions. We detail the architecture, including a dynamic parameter weighting system and a crucial Human in the-Loop (HITL) validation protocol, to ensure both adaptability and real-world credibility.

cs.AI↗

arXiv2025

Artificial Intelligence In Patent And Market Intelligence: A New Paradigm For Technology Scouting

This paper presents the development of an AI powered software platform that leverages advanced large language models (LLMs) to transform technology scouting and solution discovery in industrial R&D. Traditional approaches to solving complex research and development challenges are often time consuming, manually driven, and heavily dependent on domain specific expertise. These methods typically involve navigating fragmented sources such as patent repositories, commercial product catalogs, and competitor data, leading to inefficiencies and incomplete insights. The proposed platform utilizes cutting edge LLM capabilities including semantic understanding, contextual reasoning, and cross-domain knowledge extraction to interpret problem statements and retrieve high-quality, sustainable solutions. The system processes unstructured patent texts, such as claims and technical descriptions, and systematically extracts potential innovations aligned with the given problem context. These solutions are then algorithmically organized under standardized technical categories and subcategories to ensure clarity and relevance across interdisciplinary domains. In addition to patent analysis, the platform integrates commercial intelligence by identifying validated market solutions and active organizations addressing similar challenges. This combined insight sourced from both intellectual property and real world product data enables R&D teams to assess not only technical novelty but also feasibility, scalability, and sustainability. The result is a comprehensive, AI driven scouting engine that reduces manual effort, accelerates innovation cycles, and enhances decision making in complex R&D environments.

cs.AI↗

arXiv2025

Analysis of Propaganda in Tweets From Politically Biased Sources

News outlets are well known to have political associations, and many national outlets cultivate political biases to cater to different audiences. Journalists working for these news outlets have a big impact on the stories they cover. In this work, we present a methodology to analyze the role of journalists, affiliated with popular news outlets, in propagating their bias using some form of propaganda-like language. We introduce JMBX(Journalist Media Bias on X), a systematically collected and annotated dataset of 1874 tweets from Twitter (now known as X). These tweets are authored by popular journalists from 10 news outlets whose political biases range from extreme left to extreme right. We extract several insights from the data and conclude that journalists who are affiliated with outlets with extreme biases are more likely to use propaganda-like language in their writings compared to those who are affiliated with outlets with mild political leans. We compare eight different Large Language Models (LLM) by OpenAI and Google. We find that LLMs generally performs better when detecting propaganda in social media and news article compared to BERT-based model which is fine-tuned for propaganda detection. While the performance improvements of using large language models (LLMs) are significant, they come at a notable monetary and environmental cost. This study provides an analysis of both the financial costs, based on token usage, and the environmental impact, utilizing tools that estimate carbon emissions associated with LLM operations.

cs.SI↗

arXiv2025

Radical scaling: beyond our feet and fingers

Scaling laws arise and are eulogized across disciplines from natural to social sciences for providing pithy, quantitative, `scale-free', and `universal' power law relationships between two variables. On a log-log plot, the power laws display as straight lines, with a slope set by the exponent of the scaling law. In practice, a scaling relationship works only for a limited range, bookended by crossovers to other scaling laws. Leading with Taylor's oft-cited scaling law for the blast radius of an explosion against time, and by collating an unprecedented amount of datasets for laser-induced, chemical and nuclear explosions, we show distinct kinematics arise at the early and late stages. We illustrate that picking objective scales for the two axes using the transitions between regimes leads to the collapse of the data for the two regimes and their crossover, but the third regime is typically not mapped to the master curve. The objective scales permit us to abandon the arbitrarily chosen anthropocentric units of measurement, like feet for length and heart-beat for time, but the decimal system with ten digits (fingers) is still part of the picture. We show a remarkable collapse of all three regimes onto a common master curve occurs if we replace the base 10 by a dimensionless radix that combines the scales from the two crossovers. We also illustrate this approach of radical scaling for capillarity-driven pinching, coalescence and spreading of drops and bubbles, expecting such generalizations will be made for datasets across many disciplines.

cond-mat.soft↗

arXiv2025

Multiscale Insights of Domain Unfolding in Fibrin Mechanical Response

Fibrinogen, the monomeric unit of fibrin, the main constituent of blood clot, has a very complex structure. The fibrinogen builds the fibrin fiber and network through the half-staggered packing via knob-hole interaction and the $\alpha$C crosslinkers. Due to its rich structure, the elastic behavior also shows a unique nature of very high stretchability and multiple regimes in stress-strain behaviour, which is not yet fully understood. We develop an Unfolding-incorporated Coarse-Grained Polymer (UCGP) model for fibrinogen to study the effect of domain unfolding on the mechanical behavior of fibrin fiber and network. Our model captures the stretching behavior of fibrinogen as observed in AFM and all-atom simulations. We further extend our model to fibrin fiber to study the effect of molecular unfolding at the fiber and network level. We anticipate that our model will be able to account for the nonlinear mechanical behavior of crosslinked fibrin gel. It is possibly the first model of this sort to consider the precise, controllable knowledge of the effects of domain unfolding in crosslinked proteins. This model can also be used to model systems that have sacrificial bonds.

cond-mat.soft↗

arXiv2025

Developing an unfolding-incorporated coarse-grained polymer model for fibrinogen to study the mechanical behaviour

Fibrinogen is a protein found in blood that forms Fibrin polymer network to build a clot during wound healing process when there is a cut in the blood vessel. The fibrin fiber is highly stretchable and shows a complex mechanical properties. The fibrin monomer, Fibrinogen, has a very complex structure which is responsible for its unusual elastic behaviour. In this work, we focus on mechanism of unfolding of D-domain of Fibrinogen, and study its effect in the mechanical behaviour. We develop a coarse-grained (CG) bead-spring model for Fibrinogen which captures the unfolding of folded D-domains along with other necessary structural properties which affect the mechanical behaviour. The results from our unfolding-incorporated coarse-grained polymer (UCGP) model matches with the experimental results. This model has capacity to serve as the minimal unit to build a large-scale hierarchical structure of fibrin fiber and network to possibly unfold the mystery of fibrin's unusual elastic behaviour. This model can also be used for other polymers having folded domains or sacrificial bonds.

cond-mat.soft↗

arXiv2024

Bridging the Data Provenance Gap Across Text, Speech and Video

Progress in AI is driven largely by the scale and quality of training data. Despite this, there is a deficit of empirical analysis examining the attributes of well-established datasets beyond text. In this work we conduct the largest and first-of-its-kind longitudinal audit across modalities--popular text, speech, and video datasets--from their detailed sourcing trends and use restrictions to their geographical and linguistic representation. Our manual analysis covers nearly 4000 public datasets between 1990-2024, spanning 608 languages, 798 sources, 659 organizations, and 67 countries. We find that multimodal machine learning applications have overwhelmingly turned to web-crawled, synthetic, and social media platforms, such as YouTube, for their training sets, eclipsing all other sources since 2019. Secondly, tracing the chain of dataset derivations we find that while less than 33% of datasets are restrictively licensed, over 80% of the source content in widely-used text, speech, and video datasets, carry non-commercial restrictions. Finally, counter to the rising number of languages and geographies represented in public AI training datasets, our audit demonstrates measures of relative geographical and multilingual representation have failed to significantly improve their coverage since 2013. We believe the breadth of our audit enables us to empirically examine trends in data sourcing, restrictions, and Western-centricity at an ecosystem-level, and that visibility into these questions are essential to progress in responsible AI. As a contribution to ongoing improvements in dataset transparency and responsible use, we release our entire multimodal audit, allowing practitioners to trace data provenance across text, speech, and video.

cs.AI↗

arXiv2024

Rendering-Refined Stable Diffusion for Privacy Compliant Synthetic Data

Growing privacy concerns and regulations like GDPR and CCPA necessitate pseudonymization techniques that protect identity in image datasets. However, retaining utility is also essential. Traditional methods like masking and blurring degrade quality and obscure critical context, especially in human-centric images. We introduce Rendering-Refined Stable Diffusion (RefSD), a pipeline that combines 3D-rendering with Stable Diffusion, enabling prompt-based control over human attributes while preserving posture. Unlike standard diffusion models that fail to retain posture or GANs that lack realism and flexible attribute control, RefSD balances posture preservation, realism, and customization. We also propose HumanGenAI, a framework for human perception and utility evaluation. Human perception assessments reveal attribute-specific strengths and weaknesses of RefSD. Our utility experiments show that models trained on RefSD pseudonymized data outperform those trained on real data in detection tasks, with further performance gains when combining RefSD with real data. For classification tasks, we consistently observe performance improvements when using RefSD data with real data, confirming the utility of our pseudonymized data.

cs.CV↗

arXiv2024

Gravitational lensing and deflection angles of generalised Ellis-Bronnikov wormhole embedded in a warped braneworld background

We investigate null trajectories, deflection angles, and gravitational lensing in the spacetime of generalized Ellis-Bronnikov (GEB) wormholes and their embedding in a five-dimensional warped braneworld background (WGEB). The GEB geometry extends the standard Ellis-Bronnikov (EB) wormhole by introducing a steepness parameter $m \geq 2$, which controls the shape of the wormhole throat while partially improving on the violation of classical energy conditions. We compare the lensing properties of the four-dimensional GEB geometry with those of its warped five-dimensional counterpart, where the effect of the extra dimension is encoded through the parameter $\delta$, associated with the photon momentum along the extra dimension. Analytic expressions for the deflection angle are obtained in both weak and strong-lensing regimes, and the known EB results are recovered for $m = 2$. For $m > 2$, analytic approximation and numerical analysis is used where exact analytic solutions are not available. We show that the parameter $m$ leaves clear and distinguishable signatures in the deflection angle, Einstein ring radius, and image positions, while the presence of the warped extra dimension modifies the effective impact parameter and leads to a broadening of the photon sphere and lensed images.

gr-qc↗

arXiv2024

Activity Recognition on Avatar-Anonymized Datasets with Masked Differential Privacy

Privacy-preserving computer vision is an important emerging problem in machine learning and artificial intelligence. Prevalent methods tackling this problem use differential privacy (DP) or obfuscation techniques to protect the privacy of individuals. In both cases, the utility of the trained model is sacrificed heavily in this process. In this work, we present an anonymization pipeline that replaces sensitive human subjects in video datasets with synthetic avatars within context, employing a combined rendering and stable diffusion-based strategy. Additionally we propose masked differential privacy ({MaskDP}) to protect non-anonymized but privacy sensitive background information. MaskDP allows for controlling sensitive regions where differential privacy is applied, in contrast to applying DP on the entire input. This combined methodology provides strong privacy protection while minimizing the usual performance penalty of privacy preserving methods. Experiments on multiple challenging action recognition datasets demonstrate that our proposed techniques result in better utility-privacy trade-offs compared to standard differentially private training in the especially demanding $\epsilon<1$ regime.

cs.CV↗

arXiv2024

Consent in Crisis: The Rapid Decline of the AI Data Commons

General-purpose artificial intelligence (AI) systems are built on massive swathes of public web data, assembled into corpora such as C4, RefinedWeb, and Dolma. To our knowledge, we conduct the first, large-scale, longitudinal audit of the consent protocols for the web domains underlying AI training corpora. Our audit of 14,000 web domains provides an expansive view of crawlable web data and how codified data use preferences are changing over time. We observe a proliferation of AI-specific clauses to limit use, acute differences in restrictions on AI developers, as well as general inconsistencies between websites' expressed intentions in their Terms of Service and their robots.txt. We diagnose these as symptoms of ineffective web protocols, not designed to cope with the widespread re-purposing of the internet for AI. Our longitudinal analyses show that in a single year (2023-2024) there has been a rapid crescendo of data restrictions from web sources, rendering ~5%+ of all tokens in C4, or 28%+ of the most actively maintained, critical sources in C4, fully restricted from use. For Terms of Service crawling restrictions, a full 45% of C4 is now restricted. If respected or enforced, these restrictions are rapidly biasing the diversity, freshness, and scaling laws for general-purpose AI systems. We hope to illustrate the emerging crises in data consent, for both developers and creators. The foreclosure of much of the open web will impact not only commercial AI, but also non-commercial AI and academic research.

cs.CL↗

arXiv2024

Modeling drop deformations and rheology of dilute to dense emulsions

We highlight the current state-of-the-art in modeling emulsion rheology, ranging from dilute to jammed dense systems. We focus on analytical and numerical methods developed for calculating, computing, and tracking drop deformation en route to developing constitutive models for flowing emulsions. We identify material properties and dimensionless parameters, collate the small deformation theories and resulting expressions for viscometric quantities, list theoretical and numerical methods, and take stock of challenges for capturing connections between drop deformation, morphology, and rheology of emulsions. We highlight the substantial progress in providing quantitative descriptions of the rheological response using analytical theories, dimensional analysis, and powerful computational fluid dynamics to determine how macroscopic rheological properties emerge from microscopic features, including deformation and dynamics of non-interacting or interacting drops and molecular aspects that control the interfacial properties.

cond-mat.soft↗

arXiv2024

Dynamic duos: the building blocks of dimensional mechanics

Mechanics studies the relationships between space, time, and matter, which can be expressed in terms of the dimensions of length $\mathcal{L}$, time $\mathcal{T}$, and mass $\mathcal{M}$. Each dimension broadens the scope of mechanics, from geometric quantities with dimensions of the form $\mathcal{L}^x$ (like lengths or areas), to kinematic quantities of the form $\mathcal{L}^x\mathcal{T}^y$ (like speeds or accelerations), and eventually ``mass-carrying'' quantities such as mass, force, momentum, energy, action, power, viscosity, etc. These standard mechanical quantities have dimensions of the form $\mathcal{M}\mathcal{L}^x\mathcal{T}^y$, where $x$ and $y$ are integers. In this contribution, we use this dimensional structure to arrange these mass-carrying quantities into a table indexed by $x$ and $y$. Ratios of quantities in the same rows provide characteristic lengths, and in the same columns characteristic times, encompassing a great variety of physical phenomena from atomic to astronomical scales. Most generally, we show that picking duos of mechanical quantities that are neither on the same row nor column yields dynamics, where one mechanical quantity is understood as impelling motion, while the other is impeding it. The force and the mass are the prototypes of impelling and impeding factors, but many other duos are possible. This review provides a novel synthesis revealing the power of dimensional analysis, to understand processes governed by the interplay of two mechanical quantities. This elementary decomposition of space, time and motion into pairs of mechanical factors is the foundation of ``dimensional mechanics'', a method that this review wishes to promote and advance. The review is complemented by online video lectures, which initiate a discussion on the elaborate interplay of two or more mechanical quantities.

physics.gen-ph↗

arXiv2024

MedLM: Exploring Language Models for Medical Question Answering Systems

In the face of rapidly expanding online medical literature, automated systems for aggregating and summarizing information are becoming increasingly crucial for healthcare professionals and patients. Large Language Models (LLMs), with their advanced generative capabilities, have shown promise in various NLP tasks, and their potential in the healthcare domain, particularly for Closed-Book Generative QnA, is significant. However, the performance of these models in domain-specific tasks such as medical Q&A remains largely unexplored. This study aims to fill this gap by comparing the performance of general and medical-specific distilled LMs for medical Q&A. We aim to evaluate the effectiveness of fine-tuning domain-specific LMs and compare the performance of different families of Language Models. The study will address critical questions about these models' reliability, comparative performance, and effectiveness in the context of medical Q&A. The findings will provide valuable insights into the suitability of different LMs for specific applications in the medical domain.

cs.CL↗

arXiv2023

Privacy Assessment on Reconstructed Images: Are Existing Evaluation Metrics Faithful to Human Perception?

Hand-crafted image quality metrics, such as PSNR and SSIM, are commonly used to evaluate model privacy risk under reconstruction attacks. Under these metrics, reconstructed images that are determined to resemble the original one generally indicate more privacy leakage. Images determined as overall dissimilar, on the other hand, indicate higher robustness against attack. However, there is no guarantee that these metrics well reflect human opinions, which, as a judgement for model privacy leakage, are more trustworthy. In this paper, we comprehensively study the faithfulness of these hand-crafted metrics to human perception of privacy information from the reconstructed images. On 5 datasets ranging from natural images, faces, to fine-grained classes, we use 4 existing attack methods to reconstruct images from many different classification models and, for each reconstructed image, we ask multiple human annotators to assess whether this image is recognizable. Our studies reveal that the hand-crafted metrics only have a weak correlation with the human evaluation of privacy leakage and that even these metrics themselves often contradict each other. These observations suggest risks of current metrics in the community. To address this potential risk, we propose a learning-based measure called SemSim to evaluate the Semantic Similarity between the original and reconstructed images. SemSim is trained with a standard triplet loss, using an original image as an anchor, one of its recognizable reconstructed images as a positive sample, and an unrecognizable one as a negative. By training on human annotations, SemSim exhibits a greater reflection of privacy leakage on the semantic level. We show that SemSim has a significantly higher correlation with human judgment compared with existing metrics. Moreover, this strong correlation generalizes to unseen datasets, models and attack methods.

cs.CV↗

arXiv2023

Learning Tube-Certified Control using Robust Contraction Metrics

Control design for general nonlinear robotic systems with guaranteed stability and/or safety in the presence of model uncertainties is a challenging problem. Recent efforts attempt to learn a controller and a certificate (e.g., a Lyapunov function or a contraction metric) jointly using neural networks (NNs), in which model uncertainties are generally ignored during the learning process. In this paper, for nonlinear systems subject to bounded disturbances, we present a framework for jointly learning a robust nonlinear controller and a contraction metric using a novel disturbance rejection objective that certifies a tube bound using NNs for user-specified variables (e.g. control inputs). The learned controller aims to minimize the effect of disturbances on the actual trajectories of state and/or input variables from their nominal counterparts while providing certificate tubes around nominal trajectories that are guaranteed to contain actual trajectories in the presence of disturbances. Experimental results demonstrate that our framework can generate tighter (smaller) tubes and a controller that is computationally efficient to implement.

eess.SY↗

arXiv2022

A Roadmap for Greater Public Use of Privacy-Sensitive Government Data: Workshop Report

Government agencies collect and manage a wide range of ever-growing datasets. While such data has the potential to support research and evidence-based policy making, there are concerns that the dissemination of such data could infringe upon the privacy of the individuals (or organizations) from whom such data was collected. To appraise the current state of data sharing, as well as learn about opportunities for stimulating such sharing at a faster pace, a virtual workshop was held on May 21st and 26th, 2021, sponsored by the National Science Foundation (NSF) and National Institute of Standards and Technologies (NIST), and the White House Office of Science and Technology Policy (OSTP), where a multinational collection of researchers and practitioners were brought together to discuss their experiences and learn about recently developed technologies for managing privacy while sharing data. The workshop specifically focused on challenges and successes in government data sharing at various levels. The first day focused on successful examples of new technology applied to sharing of public data, including formal privacy techniques, synthetic data, and cryptographic approaches. Day two emphasized brainstorming sessions on some of the challenges and directions to address them.

cs.CR↗

arXiv2022

Geodesic Congruences in 5D Warped Ellis-Bronnikov Spacetimes

We study the timelike geodesic congruences in the generalized Ellis-Bronnikov spacetime (4D-GEB) and in recently proposed 5D model where a 4D-GEB is embedded in a warped geometry (5D-WGEB) and conduct a comparative study. Analytical expressions of ESR variables (for 4D geometries) are found which reveal the role of the wormhole parameter. In more general 4D and 5D scenarios geodesic equation, geodesic deviation equation and Raychaudhury equations are solved numerically. The evolution of cross-sectional area of the congruences of timelike geodesics (orthogonal to the geodesic flow lines) projected on 2D-surfaces yield an interesting perspective and shows the effects of the wormhole parameter and growing/decaying warp factors. Presence of warping factor triggers rotation or accretion even in the absence of initial congruence rotation. Presence of rotation in the congruence is also found to be playing a crucial role which we discuss in detail.

gr-qc↗