Selected work

Representative Papers

Foundation-Assisted Active Learning for Object Detection Annotation

Jul 18, 2026

This work addresses key challenges in remote sensing object detection, including high annotation costs, substantial localization noise during cold-start phases, and the difficulty of existing active learning methods in disentangling localization and classification uncertainties. To overcome these issues, the authors propose a foundation model–assisted active learning and semi-automatic annotation framework that fuses a reference localization source (SA-source, built upon UPN+SAM2) with a detector prediction source (OD-source) to jointly model localization consistency and classification confidence. The approach incorporates object-level features to enable diversity-aware sampling and suppress geometric noise, and introduces a dual-source bounding box switching mechanism to refine the annotation process. Experiments on DIOR, HRSC2016, DOTAv2, and FAIR1M demonstrate that the method significantly improves sample efficiency in cold-start scenarios and enhances detection performance under low annotation budgets.

0 citationsRead paper

Cognitive Warfare: Definition, Framework, and Case Study

Mar 05, 2026

This study addresses the lack of a unified definition and effective evaluation methods in cognitive warfare, which is often oversimplified as a subset of information operations and thus fails to adequately capture the dynamics of adversarial interaction and the mechanisms of cognitive advantage. To resolve this, the work proposes a precise definition of cognitive warfare and systematically distinguishes it from information operations for the first time. It further develops an analytical framework grounded in the OODA (Observe, Orient, Decide, Act) loop, integrating cognitive modeling with case-based wargaming. This framework enables the quantification of key attributes of cognitive advantage and provides joint force commanders and analysts with an operational toolset for understanding, comparing, and evaluating cognitive warfare, thereby significantly enhancing the capacity to assess effectiveness in the cognitive domain.

0 citationsRead paper

Nonlinear Covariance Shrinkage for Hotelling's $T^2$ in High Dimension

Feb 04, 2025

This paper addresses the reduced statistical power of Hotelling’s $T^2$ test in high-dimensional, low-sample-size settings ($p/n o gamma > 0$), where inaccurate covariance matrix estimation severely impairs performance. We propose a nonlinear covariance shrinkage estimator that does not require assumptions on spiked structure or bounded condition number. Leveraging variational inference and a novel local random matrix theory, our method constructs an adaptive, eigenvector-preserving shrinkage estimator capable of accommodating arbitrary spectral shapes and broad rank regimes. Unlike conventional linear shrinkage and existing nonlinear approaches, our estimator is theoretically grounded and provably enhances detection power for mean vector testing. Extensive simulations and real-data analyses—from finance to genomics—demonstrate its robustness and superior performance over state-of-the-art alternatives.

0 citationsRead paper

Recent publications

Latest Papers

Foundation-Assisted Active Learning for Object Detection Annotation

Jul 18, 2026

This work addresses key challenges in remote sensing object detection, including high annotation costs, substantial localization noise during cold-start phases, and the difficulty of existing active learning methods in disentangling localization and classification uncertainties. To overcome these issues, the authors propose a foundation model–assisted active learning and semi-automatic annotation framework that fuses a reference localization source (SA-source, built upon UPN+SAM2) with a detector prediction source (OD-source) to jointly model localization consistency and classification confidence. The approach incorporates object-level features to enable diversity-aware sampling and suppress geometric noise, and introduces a dual-source bounding box switching mechanism to refine the annotation process. Experiments on DIOR, HRSC2016, DOTAv2, and FAIR1M demonstrate that the method significantly improves sample efficiency in cold-start scenarios and enhances detection performance under low annotation budgets.

0 citationsRead paper

Cognitive Warfare: Definition, Framework, and Case Study

Mar 05, 2026

This study addresses the lack of a unified definition and effective evaluation methods in cognitive warfare, which is often oversimplified as a subset of information operations and thus fails to adequately capture the dynamics of adversarial interaction and the mechanisms of cognitive advantage. To resolve this, the work proposes a precise definition of cognitive warfare and systematically distinguishes it from information operations for the first time. It further develops an analytical framework grounded in the OODA (Observe, Orient, Decide, Act) loop, integrating cognitive modeling with case-based wargaming. This framework enables the quantification of key attributes of cognitive advantage and provides joint force commanders and analysts with an operational toolset for understanding, comparing, and evaluating cognitive warfare, thereby significantly enhancing the capacity to assess effectiveness in the cognitive domain.

0 citationsRead paper

Nonlinear Covariance Shrinkage for Hotelling's $T^2$ in High Dimension

Feb 04, 2025

This paper addresses the reduced statistical power of Hotelling’s $T^2$ test in high-dimensional, low-sample-size settings ($p/n o gamma > 0$), where inaccurate covariance matrix estimation severely impairs performance. We propose a nonlinear covariance shrinkage estimator that does not require assumptions on spiked structure or bounded condition number. Leveraging variational inference and a novel local random matrix theory, our method constructs an adaptive, eigenvector-preserving shrinkage estimator capable of accommodating arbitrary spectral shapes and broad rank regimes. Unlike conventional linear shrinkage and existing nonlinear approaches, our estimator is theoretically grounded and provably enhances detection power for mean vector testing. Extensive simulations and real-data analyses—from finance to genomics—demonstrate its robustness and superior performance over state-of-the-art alternatives.

0 citationsRead paper