Filtreler

2022 - 20263
Filtreleri Sıfırla

Ayarlar

Bölüm: Işık Üniversitesi, Mühendislik ve Doğa Bilimleri Fakültesi, Bilgisayar Mühendisliği Bölümü × Konu: Benchmarking ×

Arama Sonuçları

Listeleniyor 1 - 3 / 3
  • Küçük Resim
    Yayın
    Analysis of single image super resolution models
    (IEEE, 2022-11-18) Köprülü, Mertali; Eskil, Mustafa Taner
    Image Super-Resolution (SR) is a set of image processing techniques which improve the resolution of images and videos. Deep learning approaches have made remarkable improvement in image super-resolution in recent years. This article aims and seeks to provide a comprehensive analysis on recent advances of models which has been used in image superresolution. This study has been investigated over other essential topics of current model problems, such as publicly accessible benchmark data-sets and performance evaluation measures. Finally, The study concluded these analysis by highlighting several weaknesses of existing base models as their feeding strategy and approved that the training technique which is Blind Feeding, which led several model to achieve state-of-the art.
  • Küçük Resim
    Yayın
    TurkEmbed: Turkish embedding model on natural language inference & sentence text similarity tasks
    (Institute of Electrical and Electronics Engineers Inc., 2025) Ezerceli, Özay; Gümüşçekiçci, Gizem; Erkoç, Tuğba; Özenç, Berke
    This paper introduces TurkEmbed, a novel Turkish language embedding model designed to outperform existing models, particularly in Natural Language Inference (NLI) and Semantic Textual Similarity (STS) tasks. Current Turkish embedding models often rely on machine-translated datasets, potentially limiting their accuracy and semantic understanding. TurkEmbed utilizes a combination of diverse datasets and advanced training techniques, including matryoshka representation learning, to achieve more robust and accurate embeddings. This approach enables the model to adapt to various resource-constrained environments, offering faster encoding capabilities. Our evaluation on the Turkish STS-b-TR dataset, using Pearson and Spearman correlation metrics, demonstrates significant improvements in semantic similarity tasks. Furthermore, TurkEmbed surpasses the current state-of-the-art model, Emrecan, on All-NLI-TR and STS-b-TR benchmarks, achieving a 1-4% improvement. TurkEmbed promises to enhance the Turkish NLP ecosystem by providing a more nuanced understanding of language and facilitating advancements in downstream applications.
  • Küçük Resim
    Yayın
    From policy to practice: a sector-agnostic operational framework for post-quantum cryptography transition
    (Institute of Electrical and Electronics Engineers Inc., 2026-03-02) Birgin, Berat; Çeliktaş, Barış
    The pace of quantum computing development necessitates not only the adoption of post-quantum cryptographic algorithms, but also the establishment of an executable and auditable institutional transition process. Although guidance documents published by the National Institute of Standards and Technology (NIST) and roadmaps proposed by the Post-Quantum Cryptography Coalition (PQCC) articulate strategic objectives, they largely remain procedural constructs lacking a concrete operational execution model. This paper presents an industry-neutral operational framework that translates policy-level post-quantum cryptography (PQC) guidance into deterministic, proof-producing process flows encompassing cryptographic asset discovery, classification, risk modeling, algorithm selection, deployment, monitoring, and governance enforcement. Central to the framework is a deterministic Quantum Risk Scoring (QRS) function, calibrated using the Analytical Hierarchy Process (AHP), which enables reproducible asset prioritization and policy-driven enforcement decisions. Framework executability is further strengthened through cryptography-aware continuous integration/continuous deployment (CI/CD) validation gates and downgrade protection mechanisms, ensuring the generation of verifiable and immutable audit artifacts. A scenario-based operational validation, implemented using open-source toolchains, demonstrates the framework’s operability, auditability, and governance alignment without relying on empirical cryptographic performance benchmarks, confirming that PQC transition can be operationalized as a verifiable lifecycle process bridging policy guidance with enforceable technical actions. Rather than introducing new cryptographic primitives, this work formalizes PQC transition as an operational systems-engineering problem centered on governance-enforced execution and lifecycle verifiability.