Large Language Models (LLMs) have shown remarkable capabilities in knowledge-intensive tasks, while they remain vulnerable when encountering misinformation. Existing studies have explored the role of LLMs in combating misinformation, but there is still a lack of fine-grained analysis on the specific aspects and extent to which LLMs are influenced by misinformation. To bridge this gap, we present MISBENCH, the current largest and most comprehensive benchmark for evaluating LLMs' behavior and knowledge preference toward misinformation. MISBENCH consists of 10,346,712 pieces of misinformation, which uniquely considers both knowledge-based conflicts and stylistic variations in misinformation. Empirical results reveal that while LLMs demonstrate comparable abilities in discerning misinformation, they still remain susceptible to knowledge conflicts and stylistic variations. Based on these findings, we further propose a novel approach called Reconstruct to Discriminate (RtD) to strengthen LLMs' ability to detect misinformation. Our study provides valuable insights into LLMs' interactions with misinformation, and we believe MISBENCH can serve as an effective benchmark for evaluating LLM-based detectors and enhancing their reliability in real-world applications. Codes and data are available at: https://github.com/GKNL/MisBench.

Despite significant advancements in Large Language Models (LLMs), developing advanced reasoning capabilities in LLMs remains a key challenge. Process Reward Models (PRMs) have demonstrated exceptional promise in enhancing reasoning by providing step-wise feedback, particularly in the context of mathematical reasoning. However, their application to broader reasoning domains remains understudied, largely due to the high costs associated with manually creating step-level supervision. In this work, we explore the potential of PRMs in graph reasoning problems - a domain that demands sophisticated multi-step reasoning and offers opportunities for automated step-level data generation using established graph algorithms. We introduce GraphSilo, the largest dataset for graph reasoning problems with fine-grained step-wise label, built using automated Task-oriented Trajectories and Monte Carlo Tree Search (MCTS) to generate detailed reasoning steps with step-wise labels. Building upon this dataset, we train GraphPRM, the first PRM designed for graph reasoning problems, and evaluate its effectiveness in two key settings: inference-time scaling and reinforcement learning via Direct Preference Optimization (DPO). Experimental results show that GraphPRM significantly improves LLM performance across 13 graph reasoning tasks, delivering a 9% gain for Qwen2.5-7B and demonstrating transferability to new graph reasoning datasets and new reasoning domains like mathematical problem-solving. Notably, GraphPRM enhances LLM performance on GSM8K and MATH500, underscoring the cross-domain applicability of graph-based reasoning rewards. Our findings highlight the potential of PRMs in advancing reasoning across diverse domains, paving the way for more versatile and effective LLMs.

In inductive link prediction for evolving knowledge graphs (KGs), interpretability is crucial yet often overlooked in relational message aggregation methods. Previous approaches typically neglect deeper relational insights, limiting their explanatory power. In this paper, we propose CAPM (Constrained Aggregation and Predictive Modeling) to address this critical gap by uniquely incorporating semantic descriptions of entities. This integration allows for a clearer understanding of how and why certain links are predicted and strengthens its ability to contextualize and clarify the relationships within the KG. In particular, CAPM combines entity type and an attention mechanism during aggregation, ensuring a sophisticated blend of structured and semantic information. This method also improves relevance assessment in relational paths, leveraging prior knowledge in adjacent relations. Tested on sparse KGs, CAPM demonstrates exceptional performance in inductive link prediction scenarios. Ablation studies confirm its superiority, particularly when combined with embedding-based methods for entity-type representation, highlighting its effectiveness in evolving KGs. Through this innovative approach, CAPM offers a comprehensive solution that balances the dynamic nature of KGs with the essential need for interpretability in link prediction.

Link prediction (LP) in knowledge graph (KG) is a crucial task that has received increasing attention recently. Due to the heterogeneous structures of KGs, various application scenarios, and demand-specific downstream objectives, there exist multiple subtasks in LP. Most studies only focus on designing a dedicated architecture for a specific subtask, which results in various complicated LP models. The isolated architectures and chaotic situations make it significant to construct a unified model that can handle multiple LP subtasks simultaneously. However, unifying all subtasks in LP presents numerous challenges, including unified input forms, task-specific context modeling, and topological information encoding. To address these challenges, we propose a topology-aware generative framework, namely UniLP, which utilizes a generative pre-trained language model to accomplish different LP subtasks universally. Specifically, we introduce a context demonstration template to convert task-specific context into a unified generative formulation. Based on the unified formulation, to address the limitation of transformer architecture that may overlook important structural signals in KGs, we design novel topology-aware soft prompts to deeply couple topology and text information in a contextualized manner. Extensive experiment results demonstrate that our framework achieves substantial performance gain and provides a real unified end-to-end solution for the whole LP subtasks. We also perform comprehensive ablation studies to support in-depth analysis of each component in UniLP.

In wireless sensor networks, data aggregation protocols are used to prolong the network lifetime. However, the problem of how to perform data aggregation while preserving data privacy is challenging. This paper presents a polynomial regression-based data aggregation protocol that preserves the privacy of sensor data. In the proposed protocol, sensor nodes represent their data as polynomial functions to reduce the amount of data transmission. In order to protect data privacy, sensor nodes secretly send coefficients of the polynomial functions to data aggregators instead of their original data. Data aggregation is performed on the basis of the concealed polynomial coefficients, and the base station is able to extract a good approximation of the network data from the aggregation result. The security analysis and simulation results show that the proposed scheme is able to reduce the amount of data transmission in the network while preserving data privacy. Copyright © 2013 John Wiley & Sons, Ltd.

This paper investigates the fundamental performance limits of medium access control (MAC) protocols for particular multihop, RF-based wireless sensor networks and underwater sensor networks. A key aspect of this study is the modeling of a fair-access criterion that requires sensors to have an equal rate of underwater frame delivery to the base station. Tight upper bounds on network utilization and tight lower bounds on the minimum time between samples are derived for fixed linear and grid topologies. The significance of these bounds is two-fold: First, they hold for any MAC protocol under both single-channel and half-duplex radios; second, they are provably tight. For underwater sensor networks, under certain conditions, we derive a tight upper bound on network utilization and demonstrate a significant fact that the utilization in networks with propagation delay is larger than that in networks with no propagation delay. The challenge of this work about underwater sensor networks lies in the fact that the propagation delay impact on underwater sensor networks is difficult to model. Finally, we explore bounds in networks with more complex topologies.

This paper explores the segmentation of a monitoring space generated by binary pyroelectric sensors and reference structure. Each segment in the monitoring space can be identified by a state of a family of binary sensors, called a signature. In this paper, we show that the maximum number of signatures in a sensor network with n binary sensors is 2nand that it can be achieved by n modulators under a defined procedure and one assumption. Furthermore, we prove that the maximum number of signatures can be achieved in a sensor network with n binary sensors without the constraint of the number of modulators. Finally, we explore the signature combination for several simple cases.

This paper investigates fundamental performance limits of medium access control (MAC) protocols for particular underwater multi-hop sensor networks under a fair-access criterion requiring that sensors have an equal rate of underwater frame delivery to a base station. Tight upper bounds on network utilization and tight lower bounds on minimum time between samples are derived for fixed linear topology. The paper also examines the implication of the end-to-end performance bounds regarding the traffic rate and sensing time interval of individual sensors.