arXiv subjects

Ligang Sun

Publications and source records attributed to Ligang Sun.

3 recordsLinked to original sources

arXiv2025

RobustSora: De-Watermarked Benchmark for Robust AI-Generated Video Detection

The proliferation of AI-generated video models poses new challenges to information integrity and digital trust. A key confound, however, remains unaddressed: commercial generators embed visible overlay watermarks for provenance tracking, yet no existing benchmark controls for this variable, leaving open whether detectors learn genuine generation artefacts or merely associate watermark patterns with AI-generated labels. We present RobustSora, a benchmark of 6,500 manually verified videos in four categories: Authentic-Clean (A-C), Generated-Watermarked (G-W), Generated-DeWatermarked (G-DeW), and Authentic-Spoofed (A-S), sourced from Vript, DVF, and UltraVideo (authentic) and from Sora, Sora 2, Pika, Open-Sora 2, and KLing (generated). Two evaluation tasks isolate watermark effects: Task-I (Watermark Erasure Robustness) tests detection on watermark-removed AI videos; Task-II (Watermark Spoofing Robustness) measures false-alarm rates on authentic videos injected with fake watermarks. Across ten models spanning specialized detectors, transformer classifiers, and MLLMs, watermark manipulation induces accuracy changes of $-9.4$ to $+1.6$ pp (mean 6.6 pp; $p{<}0.01$ for 7/10 models on each task). A placebo control bounds inpainting-artefact confounds at $\le$2 pp, and a watermark-aware training augmentation recovers 3-4 pp on both tasks, together providing causal evidence that detectors actively rely on watermark cues. Per-generator breakdown shows that Sora 2 induces drops of $-11$ to $-14$ pp versus $-3$ to $-6$ pp for Pika and Open-Sora 2, indicating that watermark prominence, rather than detector architecture, is the principal driver of dependency. These results argue for watermark-aware evaluation and training in AIGC video detection. Dataset, evaluation code, and pretrained checkpoints will be released.

cs.CV↗

arXiv2023

General-purpose machine-learned potential for 16 elemental metals and their alloys

Machine-learned potentials (MLPs) have exhibited remarkable accuracy, yet the lack of general-purpose MLPs for a broad spectrum of elements and their alloys limits their applicability. Here, we present a feasible approach for constructing a unified general-purpose MLP for numerous elements, demonstrated through a model (UNEP-v1) for 16 elemental metals and their alloys. To achieve a complete representation of the chemical space, we show, via principal component analysis and diverse test datasets, that employing one-component and two-component systems suffices. Our unified UNEP-v1 model exhibits superior performance across various physical properties compared to a widely used embedded-atom method potential, while maintaining remarkable efficiency. We demonstrate our approach's effectiveness through reproducing experimentally observed chemical order and stable phases, and large-scale simulations of plasticity and primary radiation damage in MoTaVW alloys. This work represents a significant leap towards a unified general-purpose MLP encompassing the periodic table, with profound implications for materials science.

cond-mat.mtrl-sci↗

arXiv2018

A New Kalman Filter Model for Nonlinear Systems Based on Ellipsoidal Bounding

In this paper, a new filter model called set-membership Kalman filter for nonlinear state estimation problems was designed, where both random and unknown but bounded uncertainties were considered simultaneously in the discrete-time system. The main loop of this algorithm includes one prediction step and one correction step with measurement information, and the key part in each loop is to solve an optimization problem. The solution of the optimization problem produces the optimal estimation for the state, which is bounded by ellipsoids. The new filter was applied on a highly nonlinear benchmark example and a two-dimensional simulated trajectory estimation problem, in which the new filter behaved better compared with extended Kalman filter results. Sensitivity of the algorithm was discussed in the end.

math.OC↗