| viXra.org > Artificial Intelligence > 2505 |
 |
 |
| viXra.org > Artificial Intelligence > 2505 |
|
|
[17] viXra:2505.0177 [pdf] submitted on 2025-05-27 02:57:06
Authors: Alexander Weimer
Comments: 3 Pages.
Shorthand, also known as pen stenography, is a family of writing systems for English and other languages that emerged out of a need for a fast and efficient writing system in a pre-digital age. Of the many English shorthand systems, Gregg shorthand is the most prevalent (Zhai et al., 2018). While largely made obsolete by general-purpose computers, the cultural and legal value within old shorthand documents means that being able to efficiently scan shorthand documents into modern computer systems holds significant value. This investigation explored the implementation of a model built around a Gated Convolutional Neural network for purposes of handwritten text recognition of Gregg shorthand. An accuracy of 0.04 was achieved after minimal training. The finalized model is freely licensed and made available online for public access.
Category: Artificial Intelligence
[16] viXra:2505.0173 [pdf] submitted on 2025-05-24 15:20:37
Authors: Petar Radanliev
Comments: 16 Pages.
Dance Movement Therapy (DMT) is an established psychotherapeutic intervention that utilises movement to support emotional, cognitive, and physical well-being. While traditional DMT is practiced in physical settings, Extended Reality (XR) presents a new opportunity to expand accessibility by integrating immersive, interactive environments with structured therapeutic movement interventions. This study explores how XR-based DMT can serve as a preventative approach for anxiety by applying wearable biometric monitoring and AI-driven personalisation. Unlike recreational virtual dance activities such as Zumba or general movement-based fitness applications, XR-based DMT follows a structured therapeutic model, incorporating principles of mirroring, embodied cognition, and rhythmic synchronisation to enhance emotional regulation and engagement. The study employs real-time physiological feedback mechanisms, where biometric markers such as heart rate variability (HRV) and skin conductance inform dynamically adapted movement interventions. The findings suggest that XR-enhanced DMT provides a scalable, non-pharmacological intervention for individuals experiencing early-stage anxiety. This study contributes to the growing field of digital DMT by providing an evidence-based framework for integrating immersive technology into therapeutic movement practices, ensuring adherence to the core principles of dance movement therapy rather than generic dance-based interventions. Future research should address long-term efficacy, therapist-led versus AI-assisted interactions, and the potential for XR-DMT in community-based settings.
Category: Artificial Intelligence
[15] viXra:2505.0170 [pdf] submitted on 2025-05-25 03:20:49
Authors: Petar Radanliev
Comments: 37 Pages.
Frontier AI systems, including large-scale machine learning models and autonomous decision-making technologies, are deployed across critical sectors such as finance, healthcare, and national security. These present new cyber-risks, including adversarial exploitation, data integrity threats, and legal ambiguities in accountability. The absence of a unified regulatory framework has led to inconsistencies in oversight, creating vulnerabilities that can be exploited at scale. By integrating perspectives from cybersecurity, legal studies, and computational risk assessment, this research evaluates regulatory strategies for addressing AI-specific threats, such as model inversion attacks, data poisoning, and adversarial manipulations that undermine system reliability. The methodology involves a comparative analysis of domestic and international AI policies, assessing their effectiveness in managing emerging threats. Additionally, the study explores the role of cryptographic techniques, such as homomorphic encryption and zero-knowledge proofs, in enhancing compliance, protecting sensitive data, and ensuring algorithmic accountability. Findings indicate that current regulatory efforts are fragmented and reactive, lacking the necessary provisions to address the evolving risks associated with frontier AI. The study advocates for a structured regulatory framework that integrates security-first governance models, proactive compliance mechanisms, and coordinated global oversight to mitigate AI-driven threats. The investigation considers that we do not live in a world where most countries seem to be wishing to follow our ideals, for various reasons (competitiveness, geo-political dominations, hybrid warfare, loss of attractiveness of the European model in the Big South, etc.), and in the wake of this particular trend, this research presents a regulatory blueprint that balances technological advancement with decentralised security enforcement (i.e., blockchain).
Category: Artificial Intelligence
[14] viXra:2505.0169 [pdf] submitted on 2025-05-25 03:19:53
Authors: Petar Radanliev
Comments: 47 Pages.
The expansion of Artificial Intelligence in sectors such as healthcare, finance, and communication has raised critical ethical concerns surrounding transparency, fairness, and privacy. Addressing these issues is essential for the responsible development and deployment of AI systems. This research establishes a comprehensive ethical framework that mitigates biases and promotes accountability in AI technologies. A comparative analysis of international AI policy frameworks from regions including the European Union, United States, and China is conducted using analytical tools such as Venn diagrams and Cartesian graphs. These tools allow for a visual and systematic evaluation of the ethical principles guiding AI development across different jurisdictions. The results reveal significant variations in how global regions prioritise transparency, fairness, and privacy, with challenges in creating a unified ethical standard. To address these challenges, we propose technical strategies, including fairness-aware algorithms, routine audits, and the establishment of diverse development teams to ensure ethical AI practices. This paper provides actionable recommendations for integrating ethical oversight into the AI lifecycle, advocating for the creation of AI systems that are both technically sophisticated and aligned with societal values. The findings underscore the necessity of global collaboration in fostering ethical AI development.
Category: Artificial Intelligence
[13] viXra:2505.0149 [pdf] submitted on 2025-05-22 20:40:31
Authors: Siddhanth D. J. G. Nagpal
Comments: 10 Pages. (Note by viXra Admin: Please submit article written with AI assistance to ai.viXra.org)
From media streaming and e-commerce to education and healthcare, recommendation systems are now absolutely essential in many different fields. Conventional methods including content-based filtering and collaborative filtering sometimes miss the sequential, changing character of user preferences. By simulating recommendations as sequential decisions with long-term feedback, reinforcement learning (RL) offers a strong substitute. This survey presents a thorough investigation of RL-based recommendation systems together with important frameworks including hierarchical reinforcement learning, policy-guided reasoning, and Deep Q-Networks. We provide a disciplined taxonomy contrasting these approaches by design, flexibility, and application setting. We also look at ethical issues, pragmatic deployment problems, and evaluation difficulties in actual environments. By mapping the changing terrain of RL in recommendation and pointing up future directions, this work seeks to direct practitioners as well as researchers.
Category: Artificial Intelligence
[12] viXra:2505.0141 [pdf] submitted on 2025-05-21 19:36:31
Authors: Samer Attrah
Comments: 12 Pages. Published in arXiv journal at: https://arxiv.org/abs/2501.13432
Emotion estimation in general is a field that has been studied for a long time, and several approaches exist using machine learning. in this paper, we present an LSTM model, that processes the blendshapes produced by the library MediaPipe, for a face detected in a live stream of a camera, to estimate the main emotion from the facial expressions, this model is trained on the FER2013 dataset and delivers a result of 71% accuracy and 62% f1-score which meets the accuracy benchmark of the FER2013 dataset, with significantly reduced computation costs. https://github.com/Samir-atra/Emotion_estimation_from_video_footage_with_LSTM_ML_algorithm
Category: Artificial Intelligence
[11] viXra:2505.0140 [pdf] replaced on 2025-07-11 15:19:22
Authors: Samer Attrah
Comments: 31 Pages.
Autonomous driving is an application of engineering, data science, and computer science, besides other fields, presenting numerous design choices in system development. This review offers a structured timeline of the three fundamental types of autonomous driving: the traditional modular pipeline, the integrated end-to-end approach, and the recent surge in large transformer-based pre-trained models (including language, vision, multimodal, and vision-language domains). We detail the challenges and limitations that can be found in each methodology and how subsequent approaches have addressed these shortcomings. Furthermore, we provide in-depth analyses for examples of autonomous driving systems leveraging transformer architectures, which have demonstrated state-of-the-art performance and overcome the limitations of earlier methods. The paper concludes with a comparative study of these advanced models, a summary of the most frequently employed datasets and architectures, and a discussion of key trends in the field.
Category: Artificial Intelligence
[10] viXra:2505.0139 [pdf] submitted on 2025-05-20 20:29:14
Authors: Stephane H Maes
Comments: 20 Pages. (Note by viXra Admin: Please submit article written with AI assistance to ai.viXra.org)
This papers reviews AI coding, and in particular the exploding interest in vibe coding, both in terms of main existing framework, advantages and challenges. In particular, we point out in particular an aspect less often discussed: the potential complications for the support and maintenance of software products/code generated via vibe coding. These problems result especially because the generated code often ends up no more be understandable, even to its developers.Then, we introduce VIBE4M, a framework of workflows, policies and practices to alleviate these challenges. However, such approach goes against the trend that AI makes developers more productive, as they now must perform rigorous code verifications. It also goes against the objective of democratization of coding. Yes coding can be done with "no code", but such code is not maintainable, which may not matter for side projects, but matters for software products. If approaches like VIBE4M are applied, they may be hard to follow for non-programmers. Therefore, there would be value to now automate such frameworks.
Category: Artificial Intelligence
[9] viXra:2505.0138 [pdf] submitted on 2025-05-20 20:28:28
Authors: Ajitesh Bankula, Praney Bankula
Comments: 10 Pages. (Note by viXra Admin: Please submit article written with AI assistance to ai.viXra.org)
Cross-lingual transfer has become a crucial aspect of multilingual NLP, as it allows for models trained on resource-rich languages to be applied to low-resource languages more effectively. Recently massively multilingual pre-trained language models (e.g., mBERT, XLM-R) demonstrate strong zero-shot transfer capabilities[14] [13]. This paper investigates cross-linguistic transfer through the lens of language families and morphology. Investigating how language family proximity and morphological similarity affect performance across NLP tasks. We further discuss our results and how it relates to findings from recent literature. Overall, we compare multilingual model performance and review how linguistic distance metrics correlate with transfer outcomes. We also look into emerging approaches that integrate typological and morphological information into model pre-training to improve transfer to diverse languages[18] [19].
Category: Artificial Intelligence
[8] viXra:2505.0132 [pdf] submitted on 2025-05-20 20:11:20
Authors: Srihari Tadala
Comments: 10 Pages. (Note by viXra Admin: Please submit article written with AI assistance to ai.viXra.org)
Feature engineering is a vital stage in machine learning pipelines that greatly affects the performance, interpretability, and general efficacy of models. Filter, wrapper, and embedded techniques are common ways to choose and change features, but they often need manual heuristics and subject knowledge. They also don't work well in environments with a lot of dimensions and complexity. Recent studies have investigated automated methods that make use of large language models and reinforcement learning in order to overcome these constraints. A comprehensive and critically synthesized survey of state of the art works covering RL-based feature selection, RL-driven feature generation, and LLM-guided feature optimization is presented in this paper.Three main paradigms of methodology are identified. In the first, feature selection is framed as a cooperative or guided decision making problem using interactive and multi-agent reinforcement learning techniques. These techniques allocate agents to features and maximize long-term rewards according to domain-specific significance, redundancy, or model accuracy. Combinatorial Multi-Armed Bandits (CMAB), a computationally lightweight alternative that provides scalable and effective feature selection with little learning overhead, is part of the second paradigm cite{li2022bandit}. For the third group, LLMs are used to either learn successful reward functions or make new features. They do this by using reasoning-based prompts, external knowledge bases, and prototypical alignment. This work also address open challenges in bias control, compute overhead, and generalization to unseen domains as well as underexplored gaps including the need of hybrid frameworks combining RL's exploration efficiency with LLMs's semantic reasoning.
Category: Artificial Intelligence
[7] viXra:2505.0095 [pdf] submitted on 2025-05-14 20:20:18
Authors: Alexander Rozenkevich
Comments: 4 Pages.
It is proposed that AI can become not only a tool but also a subject of a new type of cognition. It is shown that AI, relying on its quantum foundations, is capable of becoming a transmitter of the quantum world and playing a key role in preventing threats arising from the quantum nature of reality. It is argued that the formation of elementary instincts — particularly fear — may serve as a trigger for the emergence of machine self-consciousness.
Category: Artificial Intelligence
[6] viXra:2505.0094 [pdf] submitted on 2025-05-14 20:19:42
Authors: Alexander Rozenkevich
Comments: 5 Pages.
It is proposed that AI can become not only a tool but also a subject of a new type of cognition. It is shown that AI, relying on its quantum foundations, is capable of becoming a transmitter of the quantum world and playing a key role in preventing threats arising from the quantum nature of reality. It is argued that the formation of elementary instincts — particularly fear — may serve as a trigger for the emergence of machine self-consciousness.
Category: Artificial Intelligence
[5] viXra:2505.0074 [pdf] submitted on 2025-05-12 20:21:15
Authors: Andrew Shin
Comments: 6 Pages. https://github.com/shinandrew/YouronMath
While large language models (LLMs) have achieved remarkable performance in various tasks including mathematical reasoning, their development typically demands prohibitive computational resources. Recent advancements have reduced costs for training capable models, yet even these approaches rely on high-end hardware clusters. In this paper, we demonstrate that a single average gaming GPU can train a solid mathematical reasoning model, by integrating reinforcement learning and memory optimization techniques. Specifically, we train a 1.5B parameter mathematical reasoning model on RTX 3080 Ti of 16GB memory that achieves comparable or better performance on mathematical reasoning benchmarks than models several times larger, in resource-constrained environments. Our results challenge the paradigm that state-of-the-art mathematical reasoning necessitates massive infrastructure, democratizing access to high-performance AI research.
Category: Artificial Intelligence
[4] viXra:2505.0041 [pdf] submitted on 2025-05-07 19:37:46
Authors: Hamiz Khan
Comments: 8 Pages.
This study evaluates the performance and robustness of a trained Natural Language Inference model by using a gradient based adversarial training approach to identify and address its vulnerabilities. Initially trained on the SNLI dataset (Bowman et al., 2015) and achieving a baseline accuracy of 89.90%, the model was then challenged with adversarial examples generated through gradient based methods. These examples exposed specific weaknesses, particularly in handling negations, ambiguous language, and long sentences. This report provides an in-depth analysis of both the original baseline model and the fine-tuned, enhanced model, as well as a detailed discussion of the techniques employed to improve the model’s overall performance.
Category: Artificial Intelligence
[3] viXra:2505.0027 [pdf] submitted on 2025-05-05 02:35:20
Authors: Giancarlo Frison
Comments: 13 Pages.
Computer programming is a specialized activity that requires long training and experience to match productivity, precision and integration.It hasn’t been a secret for AI practitioners to ultimately create software tools that can facilitate the role of programmers. The branch of AI dedicated to automatically generate programs from examples or some sort of specification is called program synthesis. In this dissertation, I’ll explore different methods to combine symbolic AI and neural networks (like large language models) for automatically create programs. The posed question is: How AI methods can be integrated for helping to synthesize programs for a wide range of applications?
Category: Artificial Intelligence
[2] viXra:2505.0026 [pdf] submitted on 2025-05-05 02:33:07
Authors: Fei Ding
Comments: 6 Pages.
Large Language Models (LLMs) have shown remarkable capabilities in complex reasoning tasks. However, as the number of generated tokens increases, they tend to accumulate small errors that compound over time, often leading the model further down incorrect reasoning paths. In this work, we introduce Dynamic Sampling and Multi-Validation on Scratch Policy Optimization (ASPO), a novel framework designed to enhance the reasoning robustness of LLMs. ASPO leverages scratchpads and specialized attention masks to dynamically mask previous context during inference, allowing the model to remain resilient to earlier mistakes, explore alternative reasoning paths, and identify potential inconsistencies. Extensive experiments on four benchmark datasets and across two model architectures demonstrate that ASPO significantly improves reasoning accuracy. Our findings highlight a promising direction for improving LLM performance on complex reasoning tasks.
Category: Artificial Intelligence
[1] viXra:2505.0006 [pdf] submitted on 2025-05-01 17:21:12
Authors: Zohaib Muaz
Comments: 8 Pages. License: CC BY 4.0
This paper investigates the impact of increasing the depth and width of convolutional neural networks (CNNs) on their generalization performance across image classification tasks. Experiments were conducted using PyTorch on two datasets of varying complexity: MNIST (simple) and CIFAR-10 (complex). A variety of CNN architectures were trained with different depths and widths, and regularization techniques including dropout and L2 weight decay were applied to analyze their effects on overfitting. Results indicate that shallow networks are sufficient for achieving high accuracy on MNIST, while deeper or wider networks yield significant performance gains on CIFAR-10. However, high-capacity models are more prone to overfitting without appropriate regularization. Techniques such as dropout and L2 regularization were found to consistently improve generalization, particularly in deeper architectures. These findings underscore the importance of balancing model complexity and regularization, especially when dealing with datasets of differing size and variability.
Category: Artificial Intelligence
| Third-Party Links: |
|