With the proliferation of multimodal data in real-world applications, integrating time series with auxiliary modalities has become critical for accurate forecasting. Although Transformers and pre-trained language model (PLM) have enabled initial explorations of multi-domain multimodal time series analysis, several pressing challenges still remain. Specifically, coarse-grained alignment may hinder long-range semantic capture, while distribution shifts in intra-modality introduce fluctuating noise. Inspired by GNNs' capability to model spatio-temporal dependencies and contextual interactions, we propose Decoupled Dual Adaptive Temporal Graph (DDATG), a universal GNN plugin for Transformer-PLM based adaptive text-time series bimodal learning. Our framework: (1) Reconstructs global temporal patterns from decoupled local residual terms in temporal modality, enhancing local-global semantic discovery and diversifying attention mechanisms; (2) Explicitly constructs pointwise contextual connections and strengthens aggregation in textual modality, facilitating inter-modal semantic alignment. Extensive experiments across Transformer variants and domain-specific datasets demonstrate the effectiveness of DDATG. Code is available athttps://github.com/DDATG.

Image-text interleaved data, consisting of multiple images and texts arranged in a natural document format, aligns with the presentation paradigm of internet data and closely resembles human reading habits. Recent studies have shown that such data aids multimodal in-context learning and maintains the capabilities of large language models during multimodal fine-tuning. However, the limited scale and diversity of current image-text interleaved data restrict the development of multimodal large language models. In this paper, we introduce OmniCorpus, a 10 billion-scale image-text interleaved dataset. Using an efficient data engine, we filter and extract large-scale high-quality documents, which contain 8.6 billion images and 1,696 billion text tokens. Compared to counterparts (e.g., MMC4, OBELICS), our dataset 1) has 15 times larger scales while maintaining good data quality; 2) features more diverse sources, including both English and non-English websites as well as video-centric websites; 3) is more flexible, easily degradable from an image-text interleaved format to pure text corpus and image-text pairs. Through comprehensive analysis and experiments, we validate the quality, usability, and effectiveness of the proposed dataset. We hope this could provide a solid data foundation for future multimodal model research.

Efficient global Internet scanning is crucial for network measurement and security analysis. While existing target generation algorithms verify remarkable performance in largescale detection, their efficiency notably diminishes in few-seed scenarios. This decline is primarily attributed to the intricate configuration rules and sampling bias of seed addresses. Moreover, instances where BGP prefixes have few seed addresses are widespread, constituting$63.65 \%$of occurrences. We introduce 6 Vision to tackle this challenge by introducing a novel approach to encoding IPv6 addresses into images, facilitating comprehensive analysis of intricate configuration rules. Through feature stitching, 6 Vision not only improves the learnable features but also amalgamates addresses associated with configuration patterns for enhanced learning. Moreover, it integrates an environmental feedback mechanism to refine model parameters based on identified active addresses, thereby alleviating the sampling bias inherent in seed addresses. As a result, 6Vision achieves high-accuracy detection even in few-seed scenarios. The HitRate of 6 Vision is improved by$181 \% \sim 2,490 \%$compared to existing algorithms, while the CoverNum is$1.18 \sim 11.20$times that of them. Additionally, 6Vision can function as a preliminary detection module for existing algorithms, yielding a conversion gain (CG) ranging from$242 \% \sim 2,081 \%$. Ultimately, we achieve a conversion rate (CR) of$28.97 \%$for few-seed scenarios. We enrich the IPv6 hitlist, not only enhancing current target generation algorithms for large-scale address detection in few-seed scenarios but also effectively supporting IPv6 network measurement and security analysis.

Depression is a quite common mental disorder that poses serious threats to people’s physical and mental health in the modern society. Current diagnosis strategies largely rely on doctors’ experiences and patients’ cooperation, which results in a high rate of misdiagnosis in practice. It has been noticed that people with depression exhibit typical characteristics in their expressions, speech, and other aspects, which hold significant value for diagnosing. Therefore, we propose in this paper an automatic model CCFN to aid depression detection based on multimodal characteristics of human beings. With the capsule network framework, the model uses multiple capsules to extract from single modalities fine-grained features that are then aggregated into higher-level and cross-modal capsules through a dynamic routing mechanism driven by complex numbers. Such a design allows the model to adaptively acquire semantics that can discriminate depression in the cross-modal semantic space. The experiments on a standard dataset show (1) that our model outperforms existing mainstream approaches and (2) that complex-valued capsules play a key role in the success of our model.

Fast Internet-wide scanning is essential for network situational awareness and asset evaluation. However, the vast IPv6 address space makes brute-force scanning infeasible. Despite advancements in state-of-the-art methods, they do not work in seedless regions and suffer low detection efficiency and speed in regions with known active IPv6 addresses (i.e., seed addresses). Moreover, the collected active address list (i.e., IPv6 hitlist) with low coverage cannot truly represent the active IPv6 address landscape of the Internet. This paper introduces AddrMiner, a fast, efficient, and comprehensive global active IPv6 address detection system. We design a systematic active IPv6 address detection strategy that divides the IPv6 space into two detection scenarios based on the presence or absence of seed addresses to discover active IPv6 addresses from scratch and from few to many. In the seedless regions, we present AddrMiner-N, leveraging a multi-level association policy to probe active addresses. It fills the gap of address detection in seedless regions and successfully discovers active addresses in 39,899 BGP prefixes without seed addresses, with a$1.03\times $higher hit rate,$30\sim 911\times $higher speed, and$2.7\times $broader coverage, compared to existing solutions. In the regions with seed addresses, our method AddrMiner-S dynamically generates target addresses using reinforcement learning. Compared to state-of-the-art methods, AddrMiner-S achieves an impressive 56.3% hit rate and a discovery speed of 839.0/s, which is$1.9\sim 2153\times $and$1.5\sim 755\times $of existing works, respectively. Finally, we deploy AddrMiner and discover 2.1B active IPv6 addresses, including 1.7B de-aliased active addresses and 0.4B aliased addresses, through continuous probing for three years.

Abstract In order to study the application value of big data information monitoring system in early diagnosis, treatment, and nursing of human papillomavirus (HPV) infected cervical cancer, firstly, the traditional ID3‐based decision tree model was optimized through the minimum sample number of different leaf nodes. According to the optimization model, the diagnosis model of HPV‐infected cervical cancer was constructed and applied to the monitoring system of big data of cervical cancer. Diagnosis model and information monitoring system (DMIMS) was compared with Decision tree based on decision support degree (DTBDS). Then, 876 HPV‐infected cervical cancer patients diagnosed in our hospital from January 30, 2018 to January 30, 2019 were defined as the experimental group, and 670 HPV‐infected cervical cancer patients diagnosed in our hospital from January 30, 2017 to January 30, 2018 were defined as the control group. Only the experimental group used cervical cancer data and information diagnosis system. Finally, the biopsy rate after examination, the coincidence rate with pathology diagnosis, the detection rate of precancerous lesion of HPV infection, and the detection rate of HPV infection of cervical cancer were compared between the two groups. The results showed that the classification accuracy of DMIMS (95%) was higher than that of DTBDS (67%) (p < 0.05). The biopsy rate of the experimental group was significantly lower than that of the control group (p < 0.05); the coincidence rate between biopsy and pathological diagnosis, detection rate of HPV precancerous lesion, and detection rate of HPV infected cervical cancer were significantly higher in the experimental group than in the control group (p < 0.05), which showed that the big data monitoring system of cervical cancer played an important role in the detection of HPV infection cervical cancer, which can make the patients with precancerous lesions and early cervical cancer get timely diagnosis, treatment, and nursing.

A fast piecewise-defined neural network (PDNN) method is presented to produce accurate atmospheric temperature and humidity profiles from satellite hyperspectral infrared (IR) and microwave (MW) observations in all-sky conditions. The PDNN method relies on a novel classification approach and a principal component-based neural network (NN) function to better capture the nonlinear relationship between the spectral radiances and atmospheric state vectors. The algorithm was designed to only rely on satellite measurements. Large datasets were used for network training to make the retrieval robust to random errors in the reference data. For each retrieved profile, an effective quality indicator (QI) was obtained by training against the absolute value of the retrieval error. In addition, a reliable rain cloud flag was generated based on the scattering difference of cloud particles between the 50 and 118 GHz channels. Besides the independent reanalysis of field data, the algorithm’s performance was also evaluated using radiosonde measurements. Preliminary validation shows that the best temperature retrieval occurred in the mid-troposphere (around 1.0 K). Depending on the reference data, errors in the boundary layer typically ranged from 1.8 to 2.5 K. For water vapor, the retrieval error was less than 22% in the low-troposphere and less than 35% in the mid- and upper-troposphere when validated against the reanalysis field. The humidity error was approximately 10% lower when compared to radiosondes. The PDNN is used operationally to produce atmospheric temperature and moisture soundings from the Vertical Atmospheric Sounding System (VASS) of the FengYun-3E (FY-3E) satellite, an early-morning-orbit meteorological satellite launched in 2021.

This paper presents a HAO-Interaction question answering system, which exploits knowledge based question answering (KBQA) technology to quickly obtain an answer path for the input question, and then a creative text generation mechanism to acquire the final answer text. The system also provides visibility of the answer path on the user interface in order to facilitate user understanding. Different from other KBQA systems, HAO-Interaction supports users to incorporate an organizational graph database while accessing all system functionalities. In addition, the answer generation solution implemented in the system does not require any training data. HAO-Interaction keeps low response latency while ensuring a high user satisfaction. The effectiveness of HAO-Interaction has been verified by analyzing thousands of user reviews collected by the system.

In this paper, we investigate the expected outage probability and expected throughput of a multi‐user multiple‐input multiple‐output (MU‐MIMO) transmission in mobile ad hoc networks (MANET) in the presence of co‐channel interference and unpredictable inter‐beam interference. In order to achieve multi‐user diversity gain, the receiving nodes are required to report measured channel information to the transmitting node. During the time gap between channel measurement and data transmission, the channel may change with the location of moving nodes. The unpredictable behavior may cause a mismatch between the weight of beams and the instantaneous channel and the inter‐beam interference in the data transmission phase. In order to obtain the closed form expected outage probability, we categorize the behavior of nodes according to whether the inter‐beam interference exists or not and the number of received interference beams. The probability of each category and the closed form outage probability of an instantaneous MU‐MIMO transmission of each category are derived. Additionally, the expected throughput of an MU‐MIMO transmission which changes with the number of receiving nodes is obtained, and the optimal value of receiving nodes to maximize the expected throughput is discussed. Numeric results show the unpredictable inter‐beam interference degrades the outage probability performance. Reducing the duration of the time gap could improve the expected outage probability and expected throughput.

WMO Integrated Global Observing System (WIGOS) ids an integrated, coordinated and comprehensive observing system (including both surface-based and space-based components) to satisfy, in a cost-effective and sustained manner, the evolving observing requirements of Members in delivering their weather, climate, water and related environmental services. WIGOS will provide a framework for enabling the integration and optimized evolution of WMO observing systems, and of WMO's contribution to co-sponsored systems, resulting in increased knowledge and enhanced services across all WMO Programmes.