Knowledge Base Question Answering (KBQA) aims to answer natural language questions by reasoning over large-scale structured Knowledge Bases (KBs). Among existing approaches, semantic parsing-based methods have emerged as a mainstream solution, where Large Language Models (LLMs) are employed to translate questions into structured graph queries such as Logical Forms (LFs). However, this paradigm faces two critical challenges: (1) The complex semantic mapping and graph retrieval operations render direct one-shot LF generation difficult; (2) LLMs suffer from inherent hallucination issues, generating semantically plausible-seeming but factually incorrect or invalid LFs, which are non-executable. To address these challenges, this article proposes HP-Corr , a novel framework that integrates H op-wise P lanning with iterative explainable self- Corr ection for faithful knowledge reasoning. Specifically, the system utilizes a fine-tuned open source LLM for query planning and explainable self-correction. The query planner generates reasoning paths hop-by-hop, while an explainable self-correction provides hop-wise feedback, enabling interpretable path editing based on existing reasoning paths and retrieved KB knowledge. By introducing the dual-module cooperative architecture, our system performs iterative plan-then-correct to refine query paths progressively, ensuring answer reliability and LFs executability. Experimental results demonstrate significant improvements, with our approach achieving higher accuracy while substantially reducing the search space, particularly in complex multi-hop KBQA scenarios.

The large-scale knowledge base question-answering (KBQA) has become increasingly vital across various fields. In the era of large language models (LLMs), leveraging knowledge base retrieval combined with large models for knowledge reasoning has become the mainstream approach for KBQA. However, this method faces two primary challenges: (1) the high computational cost and low accuracy of similarity-based path retrieval, and (2) the relatively low accuracy of directly obtaining answers from large models. In this paper, we introduce a novel method of retrieval enhancement-stepwise reasoning (RESR), which transforms path retrieval into text semantic understanding to minimize unnecessary interference from path information in the reasoning process, guiding the LLM to generate interpretable reasoning paths rather than directly producing answers. Specifically, RESR fine-tunes the generative model through text semantic understanding to swiftly and accurately filter path information relevant to the query from a large-scale knowledge graph (KG). Additionally, we employ the Chain-of-Thought (CoT) method to guide LLMs in step-by-step reasoning, verifying the logical coherence of reasoning paths rather than directly deriving answers. Our proposed method achieves state-of-the-art (SOTA) performance on the WebQuestionsSP (WQSP) and ComplexWebQuestions (CWQ) benchmarks.

Fine-tuning pre-trained language models (PLMs) for downstream tasks has achieved remarkable success in various natural language processing (NLP) applications. To further enhance the overall performance of PLMs across different NLP tasks, we propose a novel adversarial contrastive training (ACT) method that incorporates adversarial perturbations in the parameter space. Specifically, ACT introduces adversarial perturbations to the model’s parameters, deliberately degrading its performance on downstream tasks. Subsequently, we apply contrastive learning to align the representations of specific tasks with those of the perturbed model, thereby improving the model’s robustness and generalization ability. Experimental results demonstrate that ACT significantly enhances the generalization performance of PLMs and achieves state-of-the-art results on several text classification benchmarks.

As an emerging technology, human-robot collaboration (HRC) has been implemented to enhance the performance of assembly lines and improve the safety of human workers. By integrating the advantages of human workers and collaborative robots (cobots), HRC enables production systems to process tasks consecutively, concurrently, or collaboratively. However, the introduction of cobots also makes the corresponding human-robot collaborative assembly line balancing problem more complex and difficult to solve. To solve this problem, we first propose an enhanced mixed integer program (EMIP) with various enhancement techniques and tighter bounds, and then, we develop an improved combinatorial Benders decomposition algorithm (Algorithm ICBD) with new local search strategies, Benders cuts, and acceleration procedures. To verify the effectiveness of our proposed model and algorithms, we conduct extensive computational experiments, and the results show that our proposed EMIP model is significantly better than the existing mixed integer program model; the percentages of instances that can obtain feasible and optimal solutions are increased from 82.42% to 100% and from 29.17% to 43.5%, respectively, whereas the average gap is decreased from 19.81% to 5.64%. In addition, our proposed Algorithm ICBD can get 100% of feasible solutions and 65.92% of optimal solutions for all of the test instances, and the average gap is only 1.49%. Moreover, compared with existing Benders decomposition methods for this problem, our approach yields comparatively better solutions in notably shorter average computational time when run in the same computational environment. History: Accepted by Andrea Lodi, Area Editor for Design & Analysis of Algorithms–Discrete. Funding: This research was supported by the National Natural Science Foundation Council of China [Grants 72401214, 92167206, 7221101377, 72471169, and 72231005], the Ministry of Education of China [Grant 24YJC630078], and Computation and Analytics of Complex Management Systems (Tianjin University). This research was also supported by the Tianjin Natural Science Foundation Project [Grant 23JCQNJC01900] and the Tianjin Philosophy and Social Science Planning Project [Grant TJGL21-016]. Supplemental Material: The software that supports the findings of this study is available within the paper and its Supplemental Information ( https://pubsonline.informs.org/doi/suppl/10.1287/ijoc.2023.0279 ) as well as from the IJOC GitHub software repository ( https://github.com/INFORMSJoC/2023.0279 ). The complete IJOC Software and Data Repository is available at https://informsjoc.github.io/ .

Security issues are always considered in systems with wireless networks. However, few of them investigated covert signals existing on different communication channels to confuse advisories. In this paper, we consider the cooperation between two energy‐constrained agents, who could inject covert signals. First, the system performance is measured by Kullback–Leibler divergence (KLD) to avoid much deviation. Then, the cooperative game between two agents is considered, in which two agents share the common goal at confusing advisories. More formally, this cooperative game is formulated as a Markov decision process (MDP) and the most economic strategies are obtained through reinforcement learning (RL) under the imperfect information. Finally, the feasibility of theoretical results is demonstrated on the interconnected New England test system (NETS) as well as its reduced system.