The automated comprehension of complex, multi-modal documents is fundamentally hampered by a disconnect between information extraction and reasoning. Existing systems suffer from inherent limitations. Monolithic models embed reasoning as a black box process, sacrificing transparency and depth. Meanwhile, current agent-based frameworks follow a passive, non-interactive paradigm; they handle static, global inputs rather than information derived from active exploration, which fundamentally restricts their ability to achieve structural understanding and complex reasoning. To bridge this critical gap, we introduce HEAR, a framework for Holistic Extraction and Agentic Reasoning. This innovative framework establishes a synergistic, closed-loop between a deep Vision-Language Model (VLM) driven holistic parsing engine and a collaborative multi-agent reasoning system. Our HEAR initially transforms unstructured documents into a semantically-rich, structured representation, preserving complex layouts and reconstituting multi-page tables. Subsequently, a multi-agent system performs cross-modal analysis, governed by a crucial verification protocol that forces agents to validate findings across textual and visual modalities. A conflict driven re-evaluation mechanism enables the system to dynamically re-engage the document to resolve ambiguities, thereby unifying the perception-cognition cycle. HEAR achieved first place in the ACM MM 2025 Grand Challenge on Large Vision–Language Model Learning and Applications.

Given the significant advances in Large Vision Language Models (LVLMs) in reasoning and visual understanding, mobile agents are rapidly emerging to meet users' automation needs. However, existing evaluation benchmarks are disconnected from the real world and fail to adequately address the diverse and complex requirements of users. From our extensive collection of user questionnaire, we identified five tasks: Multi-App, Vague, Interactive, Single-App, and Unethical Instructions. Around these tasks, we present MVISU-Bench, a bilingual benchmark that includes 404 tasks across 137 mobile applications. Furthermore, we propose Aider, a plug-and-play module that acts as a dynamic prompt prompter to mitigate risks and clarify user intent for mobile agents. Our Aider is easy to integrate into several frameworks and has successfully improved overall success rates by 19.55% compared to the current state-of-the-art (SOTA) on MVISU-Bench. Specifically, it achieves success rate improvements of 53.52% and 29.41% for unethical and interactive instructions, respectively. Through extensive experiments and analysis, we highlight the gap between existing mobile agents and real-world user expectations.

The live chats of developers often contain meaningful information in the form of issue–solution pairs. The issue–solution pairs can offer helpful references to others who seek solutions for the similar issues, which can improve software development efficiency by facilitating issue solving. However, previous approaches such as ISPY still struggle with unsatisfactory extraction accuracy, due to the entanglement and complexity of issue-solution pairs' feature information. To address these challenges, we propose an approach namedIS-Hunterfor mining issue-solution pairs from real-time chat data. Specifically,IS-Hunterconsists of four main components: the data preprocessing component disentangles and denoises raw chat logs, the utterance embedding component embeds utterances into vectors that subsequent components can easily process, the feature extraction component obtains textual, heuristic, and contextual feature that determines whether an utterance is topic-relevant, and the issue–solution pair prediction component predicts the utterance whether is an issue or a solution. The experimental results show that the performance of IS-Hunter outperforms the baseline methods in issue-detection and solution-extraction in terms of Precision, Recall, and F1-score. Compared with baseline methods, in issue-detection, IS-Hunter, respectively, achieves an average precision, recall, and F1-score of 0.74, 0.74, and 0.74, and it marks an obvious 4.23% improvement over the state-of-the-art approaches. Simultaneously, in solution-extraction, IS-Hunter achieves an average precision, recall, and F1-score of 0.83, 0.90, and 0.86 which is 4.88% higher than the best baseline methods.

Hypertension has become a health problem that seriously endangers human life and is the leading cause of cardiovascular disease. Many patients do not know exactly whether their blood pressure is well controlled or not, which makes their conditions worse. A context-aware intelligent system can help patients to analyse their control situation of blood pressure (BP) and provide feedback. It is especially important to determine whether the risk-factors input in the context-aware system of hypertension is appropriate. The choice of risk factors will affect the classification performance and accuracy of the system. The risk factors screening method for hypertension proposed in this paper combined the random forest algorithm and stability selection (RFSS). It can remove the redundant context information, and leave the key factors of BP control situation. Experimental results showed that the prediction accuracy achieved more than 77% prediction accuracy, and dimension of risk factors reduced by 59%. The results indicated that RFSS is an effective method in the screening of risk factors and the prediction of hypertension.

In the intelligent monitoring of the hypertensive patients, it's necessary to assess their treatment effect and give corresponding diagnostic feedback automatically. This paper proposed a hybrid decision support system (DSS) combining several data mining techniques using an improved weighted majority voting scheme (iWMV). The mass health data of hypertensive patients were used as data source of the data mining techniques, and iWMV was used to produce a proper final judgement on patients' control condition on the basis of the individual classifier results. The proposed system was trained and evaluated using data from 167 hypertensive patients. Performance analysis showed that the hybrid system could reach classification rate (CR) of 95.34% and kappa coefficient (KC) of 92.54%, much better than systems with a single classification algorithm or combining using the simple weighted majority voting scheme (WMV). Moreover, the proposed DSS showed high stability.

With the unprecedented aging of population, chronic disease become a serious problem in modern medical area. Recent advances of wireless technology make the health monitoring in home to be more convenient for the chronic disease. Reducing the influence of the interference from the other wireless equipment is one of the most important problems in wireless physical parameter monitoring device in healthcare field, as reliable data transmitting is vital in medical care. In this paper, we propose a novel interference avoidance strategy that can greatly reduce the influence of the interference from other wireless equipment to wireless physical parameter monitoring devices in the hospital or home environment. Experimental results show that the strategy has good effect on interference avoidance and reduce the package loss rate.

Recent advances of hardware and integrated chip have made the body-implanted medical device reality. In this paper, a chronic disease management system with implanted devices based on Wireless Sensor Networks (WSN) was proposed for people with chronic illnesses. We first described the architecture of the chronic disease management system. Then through analyzing the system architecture, we explained the advantages of our new system and pointed out the main issue of the short lifetime of the implanted medical devices. Then a cooperative strategy was proposed in the particular system to decrease the energy consumption, which extended the entire lifetime of all implanted devices. Finally, we evaluated the performance of our strategy with simulations.