Visual Distraction Undermines Moral Reasoning in Vision-Language Models

Moral reasoning is fundamental to safe Artificial Intelligence (AI), yet ensuring its consistency across modalities becomes critical as AI systems evolve from text-based assistants to embodied agents. Current safety techniques demonstrate success in textual contexts, but concerns remain about generalization to visual inputs. Existing moral evaluation benchmarks rely on textonly formats and lack systematic control over variables that influence moral decision-making. Here we show that visual inputs fundamentally alter moral decision-making in state-of-the-art (SOTA) Vision-Language Models (VLMs), bypassing text-based safety mechanisms. We introduce Moral Dilemma Simulation (MDS), a multimodal benchmark grounded in Moral Foundation Theory (MFT) that enables mechanistic analysis through orthogonal manipulation of visual and contextual variables. The evaluation reveals that the vision modality activates intuition-like pathways that override the more deliberate and safer reasoning patterns observed in text-only contexts. These findings expose critical fragilities where language-tuned safety filters fail to constrain visual processing, demonstrating the urgent need for multimodal safety alignment.

Topology-Preserving Deep Joint Source-Channel Coding for Semantic Communication

Many wireless vision applications, such as autonomous driving, require preservation of global structural information rather than only per-pixel fidelity. However, existing Deep joint source-channel coding (DeepJSCC) schemes mainly optimize pixel-wise losses and provide no explicit protection of connectivity or topology. This letter proposes TopoJSCC, a topology-aware DeepJSCC framework that integrates persistent-homology regularizers to end-to-end training. Specifically, we enforce topological consistency by penalizing Wasserstein distances between cubical persistence diagrams of original and reconstructed images, and between Vietoris--Rips persistence of latent features before and after the channel to promote a robust latent manifold. TopoJSCC is based on end-to-end learning and requires no side information. Experiments show improved topology preservation and peak signal-to-noise ratio (PSNR) in low signal-to-noise ratio (SNR) and bandwidth-ratio regimes.

Rethinking LLM Watermark Detection in Black-Box Settings: A Non-Intrusive Third-Party Framework

While watermarking serves as a critical mechanism for LLM provenance, existing secret-key schemes tightly couple detection with injection, requiring access to keys or provider-side scheme-specific detectors for verification. This dependency creates a fundamental barrier for real-world governance, as independent auditing becomes impossible without compromising model security or relying on the opaque claims of service providers. To resolve this dilemma, we introduce TTP-Detect, a pioneering black-box framework designed for non-intrusive, third-party watermark verification. By decoupling detection from injection, TTP-Detect reframes verification as a relative hypothesis testing problem. It employs a proxy model to amplify watermark-relevant signals and a suite of complementary relative measurements to assess the alignment of the query text with watermarked distributions. Extensive experiments across representative watermarking schemes, datasets and models demonstrate that TTP-Detect achieves superior detection performance and robustness against diverse attacks.

On Dual-Fed Pinching Antenna Systems with In-Waveguide Attenuation

Pinching antenna systems (PAS) have recently emerged as a promising architecture for flexible and reconfigurable wireless communications. However, their performance is fundamentally constrained by in-waveguide attenuation, which is non-negligible in practical dielectric waveguides and can severely degrade the achievable data rate, particularly for long waveguides. To overcome this limitation, we propose a dual-fed PAS (DF-PAS), in which each waveguide is equipped with two feed points located at the two ends, enabling dynamic feed-point selection based on user locations. This design effectively shortens the in-waveguide propagation distance and mitigates attenuation-induced power loss without modifying the waveguide structure or the PA actuation mechanism. We investigate the DF-PAS in both single- and multi-waveguide scenarios. For the single-waveguide case, we derive closed-form high-SNR approximations of the ergodic rate and obtain closed-form solutions for the optimal PA position and feed-point selection under time-division multiple access (TDMA). We then extend DF-PAS to a multi-waveguide scenario, where we first derive closed-form high-SNR approximations of the ergodic rate and then formulate a joint optimization problem over feed-point selection, PA placement, and beamforming under general orthogonal multiple access (OMA). To solve this problem efficiently, we develop a two-phase optimization framework that integrates greedy feed-point switching, gradient-based PA placement, and WMMSE-based beamforming. Simulation results demonstrate that the proposed DF-PAS consistently outperforms conventional single-fed PAS (SF-PAS) across various network configurations, validating its effectiveness as a practical and scalable solution for mitigating in-waveguide attenuation in PAS-enabled wireless networks.

Corrected Samplers for Discrete Flow Models

Discrete flow models (DFMs) have been proposed to learn the data distribution on a finite state space, offering a flexible framework as an alternative to discrete diffusion models. A line of recent work has studied samplers for discrete diffusion models, such as tau-leaping and Euler solver. However, these samplers require a large number of iterations to control discretization error, since the transition rates are frozen in time and evaluated at the initial state within each time interval. Moreover, theoretical results for these samplers often require boundedness conditions of the transition rate or they focus on a specific type of source distributions. To address those limitations, we establish non-asymptotic discretization error bounds for those samplers without any restriction on transition rates and source distributions, under the framework of discrete flow models. Furthermore, by analyzing a one-step lower bound of the Euler sampler, we propose two corrected samplers: \textit{time-corrected sampler} and \textit{location-corrected sampler}, which can reduce the discretization error of tau-leaping and Euler solver with almost no additional computational cost. We rigorously show that the location-corrected sampler has a lower iteration complexity than existing parallel samplers. We validate the effectiveness of the proposed method by demonstrating improved generation quality and reduced inference time on both simulation and text-to-image generation tasks. Code can be found in https://github.com/WanZhengyan/Corrected-Samplers-for-Discrete-Flow-Models.

Time Travel Engine: A Shared Latent Chronological Manifold Enables Historical Navigation in Large Language Models

Time functions as a fundamental dimension of human cognition, yet the mechanisms by which Large Language Models (LLMs) encode chronological progression remain opaque. We demonstrate that temporal information in their latent space is organized not as discrete clusters but as a continuous, traversable geometry. We introduce the Time Travel Engine (TTE), an interpretability-driven framework that projects diachronic linguistic patterns onto a shared chronological manifold. Unlike surface-level prompting, TTE directly modulates latent representations to induce coherent stylistic, lexical, and conceptual shifts aligned with target eras. By parameterizing diachronic evolution as a continuous manifold within the residual stream, TTE enables fluid navigation through period-specific "zeitgeists" while restricting access to future knowledge. Furthermore, experiments across diverse architectures reveal topological isomorphism between the temporal subspaces of Chinese and English-indicating that distinct languages share a universal geometric logic of historical evolution. These findings bridge historical linguistics with mechanistic interpretability, offering a novel paradigm for controlling temporal reasoning in neural networks.

Pinching Antennas in Blockage-Aware Environments: Modeling, Design, and Optimization

Pinching-antenna (PA) systems have recently emerged as a promising member of the flexible-antenna family due to their ability to dynamically establish line-of-sight (LoS) links. While most existing studies assume ideal environments without obstacles, practical indoor deployments are often obstacle-rich, where LoS blockage significantly degrades performance. This paper investigates pinching-antenna systems in blockage-aware environments by developing a deterministic model for cylinder-shaped obstacles that precisely characterizes LoS conditions without relying on stochastic approximations. Based on this model, a special case is first studied where each PA serves a single user and can only be deployed at discrete positions along the waveguide. In this case, the waveguide-user assignment is obtained via the Hungarian algorithm, and PA positions are refined using a surrogate-assisted block-coordinate search. Then, a general case is considered where each PA serves all users and can be continuously placed along the waveguide. In this case, beamforming and PA positions are jointly optimized by a weighted minimum mean square error integrated deep deterministic policy gradient (WMMSE-DDPG) approach to address non-smooth LoS transitions. Simulation results demonstrate that the proposed algorithms significantly improve system throughput and LoS connectivity compared with benchmark methods. Moreover, the results reveal that pinching-antenna systems can effectively leverage obstacles to suppress co-channel interference, converting potential blockages into performance gains.

Robust and Secure Transmission for Movable-RIS Assisted ISAC with Imperfect Sense Estimation

Reconfigurable intelligent surfaces (RISs) have been extensively applied in integrated sensing and communication (ISAC) systems due to the capability of enhancing physical layer security (PLS). However, conventional static RIS architectures lack the flexibility required for adaptive beam control in multi-user and multifunctional scenarios. To address this issue without introducing additional hardware complexity and power consumption, in this paper, we exploit a movable RIS (MRIS) architecture, which consists of a large fixed sub-surface and a smaller movable sub-surface that slides on the fixed sub-surface to achieve dynamic beam reconfiguration with static phase shifts. This paper investigates an MRIS-assisted ISAC system under imperfect sensing estimation, where dedicated radar signals serve as artificial noise to enhance secure transmission against potential eavesdroppers (Eves). The transmit beamforming vectors, MRIS phase shifts, and relative positions of the two sub-surfaces are jointly optimized to maximize the minimum secrecy rate, ensuring robust secrecy performance for the weakest user under the uncertainty of the Eves' channels. To handle the non-convexity, a convex bound is derived for the Eve channel uncertainty, and the S-procedure is employed to reformulate semi-infinite constraints as linear matrix inequalities. An efficient alternating optimization and penalty dual decomposition-based algorithm is developed. Simulation results demonstrate that the proposed MRIS architecture substantially improves secrecy performance, especially when only a small number of elements are allocated to the movable sub-surface.

From Essence to Defense: Adaptive Semantic-aware Watermarking for Embedding-as-a-Service Copyright Protection

Benefiting from the superior capabilities of large language models in natural language understanding and generation, Embeddings-as-a-Service (EaaS) has emerged as a successful commercial paradigm on the web platform. However, prior studies have revealed that EaaS is vulnerable to imitation attacks. Existing methods protect the intellectual property of EaaS through watermarking techniques, but they all ignore the most important properties of embedding: semantics, resulting in limited harmlessness and stealthiness. To this end, we propose SemMark, a novel semantic-based watermarking paradigm for EaaS copyright protection. SemMark employs locality-sensitive hashing to partition the semantic space and inject semantic-aware watermarks into specific regions, ensuring that the watermark signals remain imperceptible and diverse. In addition, we introduce the adaptive watermark weight mechanism based on the local outlier factor to preserve the original embedding distribution. Furthermore, we propose Detect-Sampling and Dimensionality-Reduction attacks and construct four scenarios to evaluate the watermarking method. Extensive experiments are conducted on four popular NLP datasets, and SemMark achieves superior verifiability, diversity, stealthiness, and harmlessness.

DualGuard: Dual-stream Large Language Model Watermarking Defense against Paraphrase and Spoofing Attack

With the rapid development of cloud-based services, large language models (LLMs) have become increasingly accessible through various web platforms. However, this accessibility has also led to growing risks of model abuse. LLM watermarking has emerged as an effective approach to mitigate such misuse and protect intellectual property. Existing watermarking algorithms, however, primarily focus on defending against paraphrase attacks while overlooking piggyback spoofing attacks, which can inject harmful content, compromise watermark reliability, and undermine trust in attribution. To address this limitation, we propose DualGuard, the first watermarking algorithm capable of defending against both paraphrase and spoofing attacks. DualGuard employs the adaptive dual-stream watermarking mechanism, in which two complementary watermark signals are dynamically injected based on the semantic content. This design enables DualGuard not only to detect but also to trace spoofing attacks, thereby ensuring reliable and trustworthy watermark detection. Extensive experiments conducted across multiple datasets and language models demonstrate that DualGuard achieves excellent detectability, robustness, traceability, and text quality, effectively advancing the state of LLM watermarking for real-world applications.