arXiv:2609.14560v1 Announce Type: new Abstract: General-purpose object detectors lose accuracy on UAV footage, where targets span only a handful of pixels and onboard compute is limited. Prior work c...
arXiv:2604.11508v3 Announce Type: replace Abstract: Fine-tuning a pretrained classifier leaves some samples reliably learned and others cycling between correct and incorrect. Curriculum learning, dat...
General-purpose object detectors lose accuracy on UAV footage, where targets span only a handful of pixels and onboard compute is limited. Prior work composes independently-validated architectural tec...
arXiv:2607. 29592v1 Announce Type: cross Abstract: The primary challenge of continual learning (CL) systems is to learn new tasks while remaining performant on previously learned tasks.
arXiv:2608. 13061v1 Announce Type: new Abstract: As the feature representation changes, replay must preserve the earlier classes.
The paper investigates the use of evidential deep learning (EDL) for multi‑modal anti‑UAV detection, comparing it with sigmoid baselines, Dempster‑Shafer evidence fusion, and uncertainty‑driven temporal sensor gating across three benchmarks (thermal tracking, RGB‑audio‑RF classification, and RGB‑IR tracking). EDL improves accuracy by up to 5.9 percentage points and better ranks classification errors, while the other components (DS fusion, Dirichlet vacuity, temporal gating) do not provide the expected benefits. The study concludes that the primary advantage of EDL stems from its training objective rather than its uncertainty estimates.
The paper introduces Cross-Scale Channel-wise Knowledge Distillation (CSCWD), a training-time framework that transfers high‑resolution spatial representations from a YOLO11m‑P2 teacher to a lightweight YOLO11n student without changing the student’s inference architecture. CSCWD aligns teacher P2 features with student P3 while also applying same‑scale distillation at deeper pyramid levels, yielding a 2.92‑point mAP@0.5 improvement over the baseline and a 2.09‑point gain over same‑scale distillation alone. In zero‑shot tests on DUT‑Anti‑UAV and on a Raspberry Pi 5, the 2.58‑million‑parameter student reaches 50.32% mAP@0.5 at 82.32 ms latency (12.15 fps) with negligible runtime or memory increase.
arXiv:2609.05762v1 Announce Type: cross Abstract: UAV image collections contain spatially and temporally related frames, yet semantic-segmentation benchmarks commonly split them at image level. Such...
arXiv:2610.07111v1 Announce Type: cross Abstract: UAVs that act as relays or base stations need accurate air-to-ground path-loss predictions for rate adaptation and placement, but propagation conditi...
arXiv:2606. 03808v1 Announce Type: cross Abstract: We propose PURGE, a machine unlearning algorithm built on a simple but an under-exploited observation: continual learning (CL) and machine unlearning (MU) which are fundamentally dual problems.
arXiv:2609.36375v1 Announce Type: new Abstract: Continual learning is usually studied through mechanisms that preserve old knowledge. We develop Successional Learning Theory (SLT), a mesoscopic accou...
arXiv:2606. 00382v1 Announce Type: new Abstract: Sequential fine-tuning of large language models forces a choice: let the shared substrate keep learning and accept catastrophic forgetting, or freeze it after task one and foreclose cross-task refinement.