Topological Necessities: Mechanism-Invariant Strategic Subgoals for Cross-Embodiment Goal-Conditioned Control
研究通过离线轨迹提取拓扑必要性子目标,解决跨实体目标条件控制问题,方法基于同调论分析成功轨迹构建门集合。
研究通过离线轨迹提取拓扑必要性子目标,解决跨实体目标条件控制问题,方法基于同调论分析成功轨迹构建门集合。
为解决灾后无人机群无线覆盖问题,提出WONDER框架,利用JEPA预测无线电影响并通过多轮协商优化轨迹选择。
Existing visual object tracking methods struggle to generalize across arbitrary modality combinations and suffer significant performance degradation when modalities are missing. To address this limitation, this work proposes AnyTrack, the first unified tracking framework capable of handling any input modality—including images, language, and audio—through a modality-aware interaction module (MIM) that enables dynamic cross-modal fusion. Additionally, a context understanding module (CUM) is introduced to model contextual information by integrating global and local cues. Evaluated on a newly constructed multimodal tracking benchmark, AnyTrack achieves state-of-the-art performance under both complete and missing modality settings, substantially enhancing model flexibility and robustness.
Rapid evolution of multimodal phishing emails challenges traditional static detection methods, which suffer from poor adaptability and limited interpretability. Method: We propose a cognition-driven dynamic defense framework that (1) constructs a unified heterogeneous email graph integrating text, metadata, and embedded resources; (2) designs a collaborative reasoning mechanism between a Cognition Graph Neural Network (CGNN) and large language models (LLMs) for deep multimodal signal fusion; and (3) introduces an adversarial self-evolution mechanism—where red teams generate evasion samples and blue teams iteratively update and compress a memory knowledge base using failure experiences—to enable continual learning and strategy refinement. Results: Evaluated on both real-world and synthetic datasets, our approach significantly outperforms state-of-the-art baselines in accuracy, generalization, and interpretability, empirically validating the efficacy of dynamic evolutionary defense paradigms.
To address the challenge of simultaneously achieving covert communication and intelligent anti-jamming under mobile, responsive jamming scenarios, this paper proposes a parallel deep reinforcement learning (DRL) framework. The method decouples the joint action space to jointly optimize covertness—via low-power spread-spectrum transmission to evade high-power tracking—and robustness—through dynamic spectrum adaptation to counter indiscriminate frequency sweeping. A novel parallel exploration-exploitation selection mechanism replaces the conventional ε-greedy policy, significantly accelerating convergence. Simulation results demonstrate that, under complex time-varying jamming, the proposed approach achieves nearly a 90% improvement in normalized throughput while concurrently enhancing both communication robustness and covertness. This work constitutes the first application of parallel DRL to mobile anti-jamming communications, establishing a new paradigm for intelligent covert communication in highly dynamic wireless environments.
研究通过离线轨迹提取拓扑必要性子目标,解决跨实体目标条件控制问题,方法基于同调论分析成功轨迹构建门集合。
为解决灾后无人机群无线覆盖问题,提出WONDER框架,利用JEPA预测无线电影响并通过多轮协商优化轨迹选择。
Existing visual object tracking methods struggle to generalize across arbitrary modality combinations and suffer significant performance degradation when modalities are missing. To address this limitation, this work proposes AnyTrack, the first unified tracking framework capable of handling any input modality—including images, language, and audio—through a modality-aware interaction module (MIM) that enables dynamic cross-modal fusion. Additionally, a context understanding module (CUM) is introduced to model contextual information by integrating global and local cues. Evaluated on a newly constructed multimodal tracking benchmark, AnyTrack achieves state-of-the-art performance under both complete and missing modality settings, substantially enhancing model flexibility and robustness.
Rapid evolution of multimodal phishing emails challenges traditional static detection methods, which suffer from poor adaptability and limited interpretability. Method: We propose a cognition-driven dynamic defense framework that (1) constructs a unified heterogeneous email graph integrating text, metadata, and embedded resources; (2) designs a collaborative reasoning mechanism between a Cognition Graph Neural Network (CGNN) and large language models (LLMs) for deep multimodal signal fusion; and (3) introduces an adversarial self-evolution mechanism—where red teams generate evasion samples and blue teams iteratively update and compress a memory knowledge base using failure experiences—to enable continual learning and strategy refinement. Results: Evaluated on both real-world and synthetic datasets, our approach significantly outperforms state-of-the-art baselines in accuracy, generalization, and interpretability, empirically validating the efficacy of dynamic evolutionary defense paradigms.
To address the challenge of simultaneously achieving covert communication and intelligent anti-jamming under mobile, responsive jamming scenarios, this paper proposes a parallel deep reinforcement learning (DRL) framework. The method decouples the joint action space to jointly optimize covertness—via low-power spread-spectrum transmission to evade high-power tracking—and robustness—through dynamic spectrum adaptation to counter indiscriminate frequency sweeping. A novel parallel exploration-exploitation selection mechanism replaces the conventional ε-greedy policy, significantly accelerating convergence. Simulation results demonstrate that, under complex time-varying jamming, the proposed approach achieves nearly a 90% improvement in normalized throughput while concurrently enhancing both communication robustness and covertness. This work constitutes the first application of parallel DRL to mobile anti-jamming communications, establishing a new paradigm for intelligent covert communication in highly dynamic wireless environments.