Global ETD Search

Search theses and dissertations gathered from participating repositories worldwide. Every result links back to the library that holds it. No account is needed.

Results

Showing 1 to 10 of 10 for “"OOD robustness"”.

  1. Out-of-distribution robustness in mission-critical computer vision applications

    … as well as in medicine. Out-of-distribution (OOD) robustness is crucial in mission-critical computer vision applications because these scenarios often involve encountering unforeseen or novel situations that may differ significantly from the training data. In mission-critical contexts, such as …

    greece Repository record for Out-of-distribution robustness in mission-critical computer vision applications (opens in a new tab)

  2. Enhancing Deep Learning Model Robustness: Insights from Out of Distribution Data Augmentation and an Innovative Image Compression Technique

    … and presenting it with out-of-distribution (OOD) data can significantly reduce accuracy, limiting real-world appli- cability. While data augmentation enhances model robustness against OOD data, there is a gap in comprehensive studies on augmentation types and their impact on OOD robustness. A …

    concordia Download PDF for Enhancing Deep Learning Model Robustness: Insights from Out of Distribution Data Augmentation and an Innovative Image Compression Technique (opens in a new tab)

  3. CRAFT: Curriculum Rank Adversarial Fine-Tuning for Robust Vision Language Models

    … results in loss of generalization and decreased robustness against distribution shifts. To address this issue, we propose Curriculum Rank Adversarial Fine-Tuning (CRAFT), a unified low-rank fine-tuning framework designed to enhance both out-of-distribution (OOD) and adversarial robustness by …

    gatech Repository record for CRAFT: Curriculum Rank Adversarial Fine-Tuning for Robust Vision Language Models (opens in a new tab)

  4. Evaluating Visual Classification Models On Out-of-distribution Shifts With Limited Training Data

    … two problem settings where out-of-distribution (OOD) classification performance is measured with limited (order of thousand images) training data -- unsupervised domain adaptation and robustness to natural distribution shifts. Overall, this thesis demonstrates that (1) such models can be better …

    gatech Repository record for Evaluating Visual Classification Models On Out-of-distribution Shifts With Limited Training Data (opens in a new tab)

  5. Data-Efficient Learning for Robust Vision and Multimodal Models in Open-World Scenarios

    … we address the challenge of out-of-distribution (OOD) robustness, which is critical for ensuring the reliability of AI systems in real-world deployments. We propose a gradient-regularized framework that leverages the local stability of score functions, combined with an energy-based clustering …

    jhu Repository record for Data-Efficient Learning for Robust Vision and Multimodal Models in Open-World Scenarios (opens in a new tab)

  6. Robust Artificial Intelligence Methods for Life Science: Graphs, Geometries, and Sequences

    … representations, handles out-of-distribution (OOD) shifts, integrates geometric and symbolic structure, and supports controllable generation. This dissertation develops scalable methods across three core modalities - graphs, 3D geometries, and sequences - while introducing two cross-cutting …

    tamu Repository record for Robust Artificial Intelligence Methods for Life Science: Graphs, Geometries, and Sequences (opens in a new tab) Embargoed until 2027-12-01

  7. Robust Artificial Intelligence Methods for Life Science: Graphs, Geometries, and Sequences

    … representations, handles out-of-distribution (OOD) shifts, integrates geometric and symbolic structure, and supports controllable generation. This dissertation develops scalable methods across three core modalities - graphs, 3D geometries, and sequences - while introducing two cross-cutting …

    tamu Repository record for Robust Artificial Intelligence Methods for Life Science: Graphs, Geometries, and Sequences (opens in a new tab) Embargoed until 2027-12-01

  8. Robust Artificial Intelligence Methods for Life Science: Graphs, Geometries, and Sequences

    … representations, handles out-of-distribution (OOD) shifts, integrates geometric and symbolic structure, and supports controllable generation. This dissertation develops scalable methods across three core modalities - graphs, 3D geometries, and sequences - while introducing two cross-cutting …

    tamu Repository record for Robust Artificial Intelligence Methods for Life Science: Graphs, Geometries, and Sequences (opens in a new tab) Embargoed until 2027-12-01

  9. Robust Artificial Intelligence Methods for Life Science: Graphs, Geometries, and Sequences

    … representations, handles out-of-distribution (OOD) shifts, integrates geometric and symbolic structure, and supports controllable generation. This dissertation develops scalable methods across three core modalities - graphs, 3D geometries, and sequences - while introducing two cross-cutting …

    tamu Repository record for Robust Artificial Intelligence Methods for Life Science: Graphs, Geometries, and Sequences (opens in a new tab) Embargoed until 2027-12-01

  10. Neural Network Models for Network Intrusion Detection: Closing the Generalisation Gap

    … that do evaluate the out-of-distribution (OOD) performance of NIDS models highlight their fragility, with results often falling to levels where they can no longer be considered operationally acceptable. The goal of this thesis is to lay the foundation for building ML-based NIDS models with …

    unsw Repository record for Neural Network Models for Network Intrusion Detection: Closing the Generalisation Gap (opens in a new tab)