Artificial Intelligence Generated Content (AIGC) assisting image production triggers controversy in journalism while attracting attention from media agencies. Key issues involve misinformation, authenticity, semantic fidelity, and interpretability. Most AIGC tools are opaque “black boxes,” hindering the dual demands of content accuracy and semantic alignment and creating ethical, sociotechnical, and trust dilemmas. This paper explores pathways for controllable image production in journalism’s special coverage and conducts two experiments with projects from China’s media agency: (1) Experiment 1 tests cross-platform adaptability via standardized prompts across three scenes, revealing disparities in semantic alignment, cultural specificity, and visual realism driven by training-corpus bias and platform-level filtering. (2) Experiment 2 builds a human-in-the-loop modular pipeline combining high-precision segmentation (SAM, GroundingDINO), semantic alignment (BrushNet), and style regulating (Style-LoRA, Prompt-to-Prompt), ensuring editorial fidelity through CLIP-based semantic scoring, NSFW/OCR/YOLO filtering, and verifiable content credentials. Traceable deployment preserves semantic representation. Consequently, we propose a human-AI collaboration mechanism for AIGC assisted image production in special coverage and recommend evaluating Character Identity Stability (CIS), Cultural Expression Accuracy (CEA), and User-Public Appropriateness (U-PA).

Flour adulteration poses significant health risks and economic losses for consumers, yet current detection methods are often impeded by high costs, limited sensitivity, and impractical laboratory requirements, leaving end users vulnerable. The difficulty is exacerbated by the need for extremely low detection limits, often at parts-per-million (ppm) levels, to identify adulterants. To address these challenges, we introduce FlourSpec, a low-cost and user-friendly system designed for on-site detection of flour adulteration at the ppm level. FlourSpec employs a spectral reconstruction algorithm to extract pertinent information from coarse-grained spectral data collected using a low-cost spectrometer. Recognizing that existing spectral reconstruction algorithms often struggle with significant errors that obscure trace adulterant information, we develop a novel end-to-end architecture that balances spectral fidelity and classification performance while minimizing the propagation of reconstruction errors. Additionally, we incorporate a hybrid attention mechanism to capture harmonic correlations within the spectra, effectively suppressing cross-band reconstruction errors. A supervised contrastive learning module is also incorporated to enhance the discriminability of trace features. Experimental evaluations demonstrate that FlourSpec achieves 97.44% accuracy in detecting multiple adulteration types and 93.12% accuracy in identifying various BPO concentrations. Notably, it is only 0.63% lower than expensive solutions and 18.74% higher than baseline systems at the same price, highlighting its effectiveness for trace-level detection.

Meat adulteration is a significant problem that can pose health risks economic losses to consumers. Current detection methods are hindered by high costs, limited capabilities, or time-consuming sample preparation, making them only accessible in laboratory tests and can not protect the safety of end-users. This paper introduces MeatSpec, a low-cost and user-friendly system for detecting meat adulteration using spectral imaging, to move the adulteration inspection out of laboratories. MeatSpec employs a multispectral camera to reduce costs while quickly capturing spectral images, but this leads to a decrease in spectral resolution and coverage. To solve this challenge, the system uses spectral reconstruction technology and innovative designs tailored for meat adulteration detection. This includes involving adulteration-related prior information during the reconstruction training phase and incorporating contrastive learning to enlarge the distances among reconstructed samples belonging to various adulteration types. Additionally, we devise distinct feature extractors for different bands based on characteristics of the reconstructed spectra and employ knowledge distillation to mitigate error in full-band reconstructed spectra while capturing features related to adulteration. Further, we extend our system to MeatSpec-G to improve its generalizability to varied adulteration conditions and unknown adulterants. To achieve this, we first propose a feature alignment-based training scheme to reduce the feature gap among samples of diverse concentrations and admixture patterns. Then, we propose a cascaded open-set recognition framework that decouples uncertainty quantification and anomaly feature discrimination, to address the limitations of softmax confidence in detecting distribution shifts and reconstruction artifacts. Experimental evaluations on 347 paired spectral images demonstrate that our system achieves a 91.06% accuracy in detecting multiple adulteration types, merely 7.78% inferior to the expensive professional solution, yet 21.58% superior to the baseline at the same price point. Moreover, our system can generalize to achieve an 88.89% detection accuracy in unknown adulteration conditions with a 27.78% improvement, and an 83.33% detection accuracy for unknown adulterants.

The emergence of multimodal large language models (MLLMs) presents promising opportunities for automation and enhancement in Electronic Design Automation (EDA). However, comprehensively evaluating these models in circuit design remains challenging due to the narrow scope of existing benchmarks. To bridge this gap, we introduce MMCircuitEval, the first multimodal benchmark specifically designed to assess MLLM performance comprehensively across diverse EDA tasks. MMCircuitEval comprises 3614 meticulously curated question-answer (QA) pairs spanning digital and analog circuits across critical EDA stages—ranging from general knowledge and specifications to front-end and back-end design. Derived from textbooks, technical question banks, datasheets, and real-world documentation, each QA pair undergoes rigorous expert review for accuracy and relevance. Our benchmark uniquely categorizes questions by design stage, circuit type, tested abilities (knowledge, comprehension, reasoning, computation), and difficulty level, enabling detailed analysis of model capabilities and limitations. Extensive evaluations reveal significant performance gaps among existing LLMs, particularly in back-end design and complex computations, highlighting the critical need for targeted training datasets and modeling approaches. MMCircuitEval provides a foundational resource for advancing MLLMs in EDA, facilitating their integration into real-world circuit design workflows. Our benchmark is available at https://github.com/cure-lab/MMCircuitEval.

Tear Film Break-Up Time (TBUT) is a critical clinical parameter in the management of dry eye disease (DED). However, traditional TBUT assessments rely on costly and time-consuming clinical procedures, while existing home-based solutions fail to provide precise TBUT values. In this work, we present Blinic, a contactless system leveraging commercial millimeter-wave (mmWave) radar to predict precise TBUT values and assess DED severity grades at home. Blinic incorporates detailed blink kinematics that are closely linked to TBUT. To address the challenge of predicting TBUT directly from radar data, we propose a teacher-student learning framework. The teacher model, trained on electronic health records (EHRs) including image-based diagnostic tests, transfers medical insights to the student model, which uses radar-captured blink dynamics. This knowledge transfer is further enhanced by a fine-tuned large language model, DryEye-LLM, which is based on clinical diagnostic reports and employs unsupervised domain adaptation to align EHRs with radar data. To ensure accurate blink motion capture, Blinic employs an antenna-coded MIMO mmWave radar design. Additionally, a query-based multitask learning module simultaneously predicts TBUT and DED severity grades, addressing potential conflicts in feature representation. Evaluated on 192 participants in collaboration with an eye clinic, Blinic demonstrates achieving a mean absolute error of 2.73 seconds for TBUT with an average accuracy of 90.54% for DED grading in real-world settings, providing a practical solution for home-based DED management.

Drones are increasingly used in applications such as last-mile delivery and infrastructure inspection, but their safe operation, especially in high-speed scenarios, remains a critical challenge. Existing vision- and LiDAR-based obstacle localization methods suffer from motion blur, latency, and low spatio-temporal resolution, making them inadequate for detecting and tracking fast-moving objects. In this work, we present HawkEye, a drone obstacle avoidance system that fuses event cameras and LiDAR to achieve high-frequency, accurate 3D tracking of dynamic objects. By leveraging the complementary strengths of both sensors, Hawkeye enables robust real-time sensing and safe evasive maneuvers, addressing a key requirement for the large-scale deployment of autonomous drones. Demo: https://wenhua00.github.io/HawkEye/.

Photovoltaic (PV) defect detection is crucial for preventing power efficiency loss and fire hazards. The industry primarily relies on the fusion of infrared and visible images for defect localization and diagnosis. However, current detection methods exhibit poor generalizability in new site environments or with altered imaging setups. While recent infrared and vision foundation models (FM) facilitate domain-invariant feature maps extraction, directly concatenating them and fine-tuning achieves limited generalizability gain to PV defect detection, due to the asymmetric dual-modal semantics of defects. In this paper, we present the first large infrared-visible model DD-LIVM to enable cross-domain defect detection. The key innovation of DD-LIVM lies in its defect-specific three-step fine-tuning strategy, which utilizes alternating modality masking. Prior to feature fusion and joint fine-tuning, the infrared and visible FM encoders are alternately masked and optimized to enhance their individual semantic utility for defect localization visibility and classification granularity, with feature distances among different defect types regulated through contrastive learning. This approach allows for the extraction of generalizable and defect-specific feature maps. Moreover, for practical employment of DD-LIVM, we propose a domain-agnostic spatial alignment algorithm for infrared-visible images before dual-modal fusion, and develop source data augmentation and adaptive detection head selection schemes based on defects' infrared characteristics to further enhance the generalizability. Extensive experiments on 7,078 dual-modal images from 9 real-world scenarios across 4 cities' PV stations demonstrate that DD-LIVM achieves an accuracy of 87.7% for cross-domain defect detection, surpassing state-of-the-art methods by 17.3%.

Monitoring sashimi freshness,i.e., histamine levels, in showcases poses a critical challenge for sushi restaurants and fresh food stores. Current histamine monitoring methods involve labor-intensive chemical experiments or expensive devices, making affordable on-site monitoring difficult. This paper proposes FreshSpec, a low-cost and automatic spectral imaging system capable of precisely monitoring histamine levels in sashimi with minimal human intervention. The low concentration of histamine, combined with the potential for other ingredients to mask its spectral characteristics, complicates precise histamine level predictions using coarse or redundant spectral data from low-cost devices. To address this issue, FreshSpec employs an innovative feature- wise spectral reconstruction (SR) framework that effectively eliminates irrelevant and redundant data while preserving critical histamine-related spectral features. Specifically, we redefine the SR reconstruction target by utilizing features derived from the encoder of the spectral foundation model that is enhanced to focus on histamine-related spectral features. Furthermore, inspired by the monotonic accumulation properties of histamine over time, we propose a histamine regression model with unsupervised continual adaptation to new sashimi samples during practical deployment. Experimental results from 240 samples of salmon, tuna, and snapper demonstrate that FreshSpec achieves an R2 of 0.9319 and an RMSE of 3.101 mg/100 g, comparable to laboratory spectral imaging systems, while outperforming baseline schemes with a 46.95% RMSE reduction and a 0.1631 R2 improvement.

To meet the new requirements of Industry 4.0, the logistics field has introduced 3D reconstruction technology. Computer vision-based solutions face challenges like bad lighting conditions and line-of-sight constraints. Meanwhile, the widespread adoption of RFID tags in supply chains offers an opportunity to enhance current reconstruction methods. In this article, we propose TagRecon, a fine-grained multi-object 3D reconstruction scheme utilizing well-deployed RFIDs. Specifically, TagRecon transforms the task of reconstruction into a problem of estimating 3D bounding boxes for tagged packages. By placing dual anchor tags on each target package, TagRecon enables accurate inference of the package’s translation and rotation using RFID-based localization and orientation sensing. Our scheme introduces a novel method to estimate rotations and translations for tagged packages, utilizing the known geometric relationship of anchor tags. Besides, to achieve simultaneous reconstruction of multiple packages, we manage to match tags from various packages through the correlation between anchor tag pairs. As far as we know, this is the first RFID-based solution that can simultaneously realize 3D translation and rotation estimation of multiple objects to a fine granularity. Experiments validate TagRecon achieves a 28.0 cm translation error and 6.8°, 6.0°, and 7.5° rotation errors for roll, pitch, and yaw angles on average.

Authenticating RFID-tagged items during mobile inventory, such as entering or leaving the warehouse, is a critical task for anti-counterfeiting. However, past authentication solutions using commercial off-the-shelf (COTS) devices cannot be applied in mobile scenarios, such as conveyors or tunnels, due to either high latency or non-robustness to tag movement. This paper introduces LoPrint, the first system to effectively authenticate mobile tagged items using the COTS orthogonal antennas existing in most infrastructures. The key insight of LoPrint is to randomly attach multiple tags on each item as a tag group and leverage the stable layout relationships of this tag group as novel fingerprints, including the relative distance matrix (RDM) and relative orientation matrix (ROM). Additionally, a new hardware fingerprint called cross-polarization ratio (CPR) is proposed to help distinguish the tag category. Furthermore, a lightweight approach is designed to robustly extract RDM, ROM, and CPR from RSSI and phase sequences under various environmental factors. LoPrint is prototyped and deployed on a conveyor in a lab environment and a tunnel in a real-world RFID warehouse, where 726 tagged items with random layouts are used for evaluation. Experimental results show that LoPrint can achieve a high authentication accuracy of 82.92% on the fixed conveyor and 79.48% on the random warehouse trolley when the size of tag group is three, outperforming the transferred stateof-the-art solution by over 10×.

Meat adulteration is a significant problem that can pose health risks economic losses to consumers. Current detection methods are hindered by high costs, limited capabilities, or time-consuming sample preparation, making them only accessible in laboratory tests and can not protect the safety of end-users. This paper introduces MeatSpec, a low-cost and user-friendly system for detecting meat adulteration using spectral imaging, to move the adulteration inspection out of laboratories. MeatSpec employs a multispectral camera to reduce costs while quickly capturing spectral images, but this leads to a decrease in spectral resolution and coverage. To solve this challenge, the system uses spectral reconstruction technology and innovative designs tailored for meat adulteration detection. This includes involving adulteration-related prior information during the reconstruction training phase and incorporating contrastive learning to enlarge the distances among reconstructed samples belonging to various adulteration types. Additionally, we devise distinct feature extractors for different bands based on characteristics of the reconstructed spectra and employ knowledge distillation to mitigate error in full-band reconstructed spectra while capturing features related to adulteration. Experimental evaluations on 347 paired spectral images demonstrate that our system achieves a 91.06% accuracy in detecting multiple adulteration types, merely 7.78% inferior to the expensive professional solution, yet 21.58% superior to the baseline at the same price point.

Unauthorized access attack has always been a critical problem in RFID systems since any illegitimate reader can conduct access commands on tags without authorization and leave no trace. Past solutions for reader authentication require either modifications on EPC-global Gen2 protocol, which are inapplicable to existing infrastructures, or numerous extra customized devices as communication monitors, which incur high overhead. In this paper, we present a universal, low-cost and effective system to authenticate RFID readers, namely ReaderPrint, which only requires an extra passive tag array and is fully compatible with Gen2 protocol. The key insight behind ReaderPrint is that the impedance mismatch degrees (IMD) of different reader antennas across channels are distinguishable. We verify this mechanism through empirical studies using vector network analyzer and further propose two brand-new forms of hardware fingerprints, i.e., IMD-induced transmission power attenuation (ITPA) and phase shifts (IPS) across channels to quantify the IMD. Besides, to address the negative impacts of environmental changes, well-refined fingerprint matching algorithms are designed accordingly. We implement a prototype of ReaderPrint and evaluate it on 96 different readers in three indoor scenarios. Experimental results show that ReaderPrint can achieve fairly high authentication accuracy of up to 97.2%, regardless of environmental or device conditions.

RFID sensing leveraging backscatter signal features (e.g., phase shift) from tags has gained increasing popularity in numerous applications but also suffers from negative impacts of environmental multipaths. Past works to address it rely on extra customized devices, labor-intensive offline training, or frequency channel hopping, all of which are non-ubiquitous or ineffective for real-life adoption. This article presents RoSense, a universal method to alleviate multipath reflections’ impacts by spinning the reader antenna, thus enabling more robust RFID sensing. Besides, RoSense requires no RF devices or offline training and operates in a nonintrusive manner. The key insight of RoSense is to exploit two properties of line-of-sight (LOS) signal when spinning the antenna, i.e., the linearity of phase changes and stability of received signal strength to attenuate the nonlinear and nonmonotonic effect of multipath signals and refine the phase shift of LOS signal. We have implemented a prototype of RoSense with COTS devices and studied two cases for evaluation: 1) material identification and 2) object localization. Experimental results show that RoSense can improve the material identification accuracy by up to 16.22% and reduce the mean localization error by up to 39.93%, greatly outperforming the state-of-the-art solutions.

RFID tags authentication is always a critical but challenging problem because only checking the EPC is vulnerable to counterfeiting attacks. Past works explore the unique backscat-ter signal features induced by tags' manufacturing imperfection as fingerprints, but fail to support simultaneous authentication for a batch of tags in practice, which is vital for large-scale RFID applications (e.g., warehouse inventory). In this paper, we present a universal architecture, namely B-AUT, to simultaneously authenticate multiple tags even with the same EPC and pinpoint them, which is fully compatible with Gen2 standard and applicable to almost all tags' hardware fingerprints proposed in existing works. The workflow of B-AUT is threefold based on our novel algorithms. First, the extracted fuzzy fingerprint and EPC are jointly exploited to cluster raw data. Second, we extract the tags' fine-grained fingerprints for genuineness validation and obtain the invalid clusters. Third, we harness localization methods to match the invalid cluster to dubious tags and further conduct small-scale re-validation to pinpoint the counterfeit tags. We have implemented a prototype of B-AUT and evaluated it in extreme cases. Experiment results demonstrate that B-AUT can maintain nearly the same authentication accuracy as that of separate authentication and reduce the time overhead by 43.3%. Moreover, the pinpointing accuracy can reach as high as 92.8%, regardless of tags' total quantities or tag models.

In radio frequency (RF) energy harvesting wireless sensor networks, employing a mobile charger (MC) is more cost-efficient and flexible to power the sensor nodes than deploying extensive stationary chargers. However, due to limited charging time of MC, the nodes' residual energy distribution (i.e., energy balance degree) after charging will directly determine the lifetime of sensor networks. To promote the energy balance degree, two important problems required to solve are: how to determine the nodes' routing scheme and how to schedule the MC to power the nodes. The two problems interact with each other. In this paper, we introduce a practical charging scenario, consider to combine the routing-based charging and charging-based routing, and propose a Jointly Balanced Routing and Charging (JBRC) scheme. Specifically, we propose a balanced routing strategy and a charging time allocation scheme, and iteratively jointly optimize them. Our goal is to achieve the energy balance in WSNs, i.e., maximizing the minimal residual energy among nodes. Additionally, the MC's moving trajectory is selected and designed. From simulation experiments, we verify the superiority of our proposed JBRC scheme.

RF Energy harvesting technology has been proved one of the effective approaches for powering battery-free wearable devices in wireless body area networks. However, excessive electromagnetic radiation is harmful to human body. In this paper, we consider real-time healthcare scenario where wearable devices worn by mobile users collect their physiological data in real time and multiple wireless chargers are deployed for energy provision. Our goal is to minimize the maximal radiation degree among mobile users while maintaining normal work of wearable devices via adaptive power control of wireless chargers. We first discrete the users' moving trajectories and transform the stubborn problem into a docile one. Then we propose a distributed algorithm with interaction of wireless chargers, wearable devices and base station to solve it. Our proposed real-time power control scheme achieves an approximation ratio of (1+e) in general case. Furthermore, one special case is discussed. Simulation results reveal that our scheme is efficient and the maximal radiation degree among users can be reduced by almost 20\% as compared to the baseline scheme.

RF energy transfer, as the main far-field wireless energy transfer technology in wireless sensor networks, allows the relatively long charging distance from wireless charger to sensor nodes. Existing charging schemes based on a mobile RF energy charger neglect the energy consumption of the charger and its limited battery capacity. Motivated by this, we consider the practical charging scenario where energy-constrained mobile charger (MC) travels along a constrained long trajectory in the network area to wirelessly power the sensors, with multiple depots (for the energy provision of MC) deployed on the trajectory to achieve high energy efficiency. In this paper, we introduce VCEC, a Velocity-Control scheme of Energy-constrained mobile Charger to maximize the minimum charged energy in nodes after MC passes through the whole trajectory. Specifically, we first simplify the initial velocity-control problem to a tractable one by discretizing the trajectory into segments and propose a distributed algorithm to solve it. Then, we present a segment merging algorithm for the real-world applications. Our VCEC scheme achieves an approximation ratio of (1-θ)(1+ ε)-1. Simulations and test-bed experiments are conducted to show that VCEC promotes the bottleneck node's charged energy by at least 20% as compared to the baseline scheme where MC moves at a constant speed.

In RF Energy harvesting wireless sensor networks, either static chargers or mobile chargers are employed to power the sensor nodes. Mobile chargers can achieve better charging performance while incurring high charging cost, which is unattractive in resource management. Motivated by this, we consider the combination of mobile chargers and static chargers to minimize the charging cost. We transform the primary problem into a tractable one and design efficient algorithms for both general scenario and on-demand scenario.

Near-field wireless power transfer has been widely used for powering sensor nodes in WSN, where mobile charger (MC) is employed to traverse the network. However, due to the limited energy of MC, MC needs to return to service station (SS) for replacing energy source for multiple times. Therefore, the location of SS directly determines the travelling distance of MC and the charging completion time as well. Motivated by this, we consider the SS positioning problem with jointly optimizing the charging tour of MC. We decompose it into two sub-problems and propose efficient algorithms to solve them based on a novel discretization method.

This paper considers the network utility maximization (NUM) problem in radio frequency (RF) energy harvesting sensor networks with static routing and mobile charging periodically, using one mobile charger (MC) to move along the constrained and fixed trajectory for cost-efficient energy provision. For the first time, the sampling rates of sensor nodes and time allocation of mobile charger are jointly optimized under energy consumption and patrolling cycle constraints, aiming to maximize the network utility. We discretize the trajectory into segments and decouple this original convex problem into two separable sub-problems and obtain the approximately optimal solution. Simulation results demonstrate that our proposed algorithm always outperforms the baseline schemes with uniform sampling rate of sensors or constant moving speed of MC.