Large Language Models (LLMs) are prone to hallucinations, producing fluent but factually incorrect statements. Recent multi-agent debate methods improve hallucination detection by jointly improving reasoning and decision-making. However, existing approaches either collaborate which amplifies shared overconfidence, or adopt adversarial preset stances, that can inject incorrect information complicating decision making. To address this, we propose Contradictory Statement Multi-Agent Debate (CSMAD), a multi-agent framework that creates structured disagreement by generating a contradictory claim for each input claim. CSMAD asks independent agents to evaluate the claim and the contradictory claim, which encourages different lines of reasoning without assigning preset stances. When the outcome is non-discriminative; both the contradictory statements are either accepted or rejected; the agents exchange rationales and update their judgments after considering opposing evidence. A final judge then decides the truth of the original claim, using both arguments as context. To make contradictory statement generation reliable, we add a Natural Language Inference (NLI) based verifier that checks whether the generated statement actually contradicts the original claim; if it does not, the system falls back to an explicit negation-based contradiction. Across public benchmarks for question answering and scientific claim verification, as well as a proprietary e-commerce claims dataset, we show that CSMAD consistently outperforms the strongest baseline for both large (Claude-3.5 Sonnet) and medium-sized (Qwen3-8B) language models, improving F1 by +2.3 and +4.1 points, respectively, while reducing LLM token cost by 28%.

Concept maps are graphical representations of the hierarchical relationships among ideas. They are frequently used to assess individual students? understanding in a given domain. We present a methodology to create a collective concept map, which illustrates a group's, not an individual's, understanding of a concept. We draw on data from 15 grades 2-12 teachers and 23 grades 6-11 students from several schools within a metropolitan area in the U.S., who were participants in a summer cybersecurity workshop for beginner learners. Using collective concept maps, we study the initial views on the topic of cybersecurity between teachers and students.

Identifying the location of randomly deployed sensor nodes in wireless sensor networks (WSNs) is a major challenge because sensor nodes have limited energy and computational capacity. This study presents a low complexity, scalable, and energy‐efficient localisation scheme capable of localising dense as well as sparse deployment of nodes using a single beacon node (SBN). The nodes act in passive listening mode to receive information transmitted by the SBN to determine their locations. The energy consumption is reduced at the nodes as well as at the SBN. The SBN has an omni‐directional antenna and a directional antenna (DA). DA can rotate to transmit beacons in different directions in the ‐plane. The time required and energy consumption for localisation are reduced at the node side as well as at the SBN side. The expressions for energy consumption and computational complexity are deducted analytically. The proposed scheme is verified with MATLAB simulation. The simulation results show a significant improvement in localisation accuracy and reduction in energy consumption.

This paper presents an energy-efficient localization of sensor nodes in wireless sensor network (WSNs) using beacons from rotating directional antenna. The localization process is scalable, fast, and easily implementable at the nodes, and works for homogeneous as well as heterogeneous networks. The process is suitable for dense and sparse deployment of nodes because communication among nodes is not required and no transmission is needed from nodes to base station (BS). Furthermore, a rotating directional antenna is used at the BS to transmit beacon signal packet containing information about the height of the antenna, current direction, transmitted power, and GPS coordinate of BS. By using the information received from the beacon packets, each node estimates its location relative to BS (in polar coordinate). The time required, energy consumption, and accuracy are estimated analytically. Simulation results show that the proposed scheme improves accuracy and energy efficiency for two-dimensional localization of sensor nodes in WSNs.