Multi-hop Knowledge Base Question Answering (KBQA) aims to find answer entities in the knowledge base that are multiple hops away from the entities in the question. Information retrieval-based (IR-based) methods extract a pivotal subgraph from the entire KB to locate candidate answers and then evaluate their plausibility through semantic matching with the question. However, we observed that the extracted subgraphs often include nodes that are weakly related or irrelevant to the question. Without a proper node filtering mechanism, the number of irrelevant nodes grows as the number of hops increases, leading to excessive consumption of computational resources. To address these challenges, this study introduces an efficient LLM-based subgraph retrieval method for multi-hop knowledge base question answering, M-ER. The framework leverages Monte Carlo Tree Search (MCTS) to transform subgraph exploration into a tree-structured search process. During the MCTS selection phase, nodes that are highly relevant to the question are prioritized for inclusion in the subgraph, eliminating the need to traverse all nodes in the KB. The framework further incorporates a large language model (LLM) to refine the search direction, ensuring that exploration remains focused on nodes relevant to the question. In addition, selected nodes are quantitatively scored, and these scores are fed back into the node selection process to effectively filter out irrelevant candidates, thereby improving the quality of the subgraph. This mechanism not only narrows the search space but also enhances the overall efficiency of multi-hop KBQA. Experiments on the WebQSP benchmark demonstrate that M-ER achieves 78.88% on the Hits@1 metric, while also improving computational efficiency. These results not only validate the effectiveness of M-ER, but also offer a viable technical path to balance performance and computational efficiency.

Text-based knowledge graph completion methods take advantage of pre-trained language models (PLM) to enhance intrinsic semantic connections of raw triplets with detailed text descriptions. Typical methods in this branch map an input query (textual descriptions associated with an entity and a relation) and its candidate entities into feature vectors, respectively, and then maximize the probability of valid triples. These methods are gaining promising performance and increasing attention for the rapid development of large language models. According to the property of the language models, the more related and specific context information the input query provides, the more discriminative the resultant embedding will be. In this paper, through observation and validation, we find a neglected fact that the relation-aware neighbors of the head entities in queries could act as effective contexts for more precise link prediction. Driven by this finding, we propose a relation-aware anchor enhanced knowledge graph completion method (RAA-KGC). Specifically, in our method, to provide a reference of what might the target entity be like, we first generate anchor entities within the relation-aware neighborhood of the head entity. Then, by pulling the query embedding towards the neighborhoods of the anchors, it is tuned to be more discriminative for target entity matching. The results of our extensive experiments not only validate the efficacy of RAA-KGC but also reveal that by integrating our relation-aware anchor enhancement strategy, the performance of current leading methods can be notably enhanced without substantial modifications.

Chain-of-thought (CoT) distillation allows a large language model (LLM) to guide a small language model (SLM) in reasoning tasks.Existing methods train the SLM to learn the long rationale in one iteration, resulting in two issues: 1) Long rationales lead to a large tokenlevel batch size during training, making gradients of core reasoning tokens (i.e., the token will directly affect the correctness of subsequent reasoning) over-smoothed as they contribute a tiny fraction of the rationale.As a result, the SLM converges to sharp minima where it fails to grasp the reasoning logic.2) The response is slow, as the SLM must generate a long rationale before reaching the answer.Therefore, we propose chunk-wise training (CWT), which uses a heuristic search to divide the rationale into internal semantically coherent chunks and focuses SLM on learning from only one chunk per iteration.In this way, CWT naturally isolates non-reasoning chunks that do not involve the core reasoning token (e.g., summary and transitional chunks) from the SLM learning for reasoning chunks, making the fraction of the core reasoning token increase in the corresponding iteration.Based on CWT, skip-thinking training (STT) is proposed.STT makes the SLM automatically skip nonreasoning medium chunks to reach the answer, improving reasoning speed while maintaining accuracy.We validate our approach on a variety of SLMs and multiple reasoning tasks.

Reasoning paths are reliable information in knowledge graph completion (KGC) in which algorithms can find strong clues of the actual relation between entities. However, in real-world applications, it is difficult to guarantee that computationally affordable paths exist toward all candidate entities. According to our observation, the prediction accuracy drops significantly when paths are absent. To make the proposed algorithm more stable against the missing path circumstances, we introduce soft reasoning paths. Concretely, a specific learnable latent path embedding is concatenated to each relation to help better model the characteristics of the corresponding paths. The combination of the relation and the corresponding learnable embedding is termed a soft path in our paper. By aligning the soft paths with the reasoning paths, a learnable embedding is guided to learn a generalized path representation of the corresponding relation. In addition, we introduce a hierarchical ranking strategy to make full use of information about the entity, relation, path, and soft path to help improve both the efficiency and accuracy of the model. Extensive experimental results illustrate that our algorithm outperforms the compared state-of-the-art algorithms by a notable margin. Our code will be released at https://github.com/7HHHHH/SRP-KGC.

Inductive relation prediction aims to predict missing connections between entities unseen during training. Recent approaches adopt binary (positive or negative) training labels, which indicate whether the query relation exists between the entities, as supervision to teach models recognizing the entity-independent relation patterns in the context (enclosed subgraph or connective path). However, we argue that in this kind of method, the trained models are guided to make relation predictions by remembering whether the query relation and its contextual relational pattern co-occur more frequently in positive or negative samples. This solution could introduce two major limitations: 1) the model struggles with long-tail combinations, i.e., the combination between query relation and the relational pattern rarely occurs during training; 2) when noisy relational patterns, which fail to provide evidence for predicting the query relation, frequently occur with the query relation in positive training samples, the model will be misled into considering the noisy relational patterns as a feature supporting the existence of the query relation. To solve these problems, we propose ToC (Thinking on Context). ToC first utilizes large language models (LLMs) to incorporate a chain of thought as an additional supervisory constraint, guiding the model to make relational predictions based on logical reasoning instead of co-occurrence frequency. Additionally, ToC employs the reasoning capabilities of LLMs to construct context-level negative samples, aiding the model in identifying and disregarding noisy relational patterns. Extensive experiments show that ToC significantly outperforms state-of-the-art methods across three widely used datasets in multiple inductive settingshttps://github.com/AI-Chen/ToC_KGC.

Chain-of-thought distillation (CoT-distillation) aims to endow small language models (SLMs) with reasoning ability to improve their performance toward specific tasks by allowing them to imitate the reasoning procedure of large language models (LLMs) beyond simply predicting the answers. Most existing CoT-distillation methods adopt a pre-thinking mechanism, allowing the SLM to generate a rationale before answering. In this way, pre-thinking enables SLM to analyze questions but makes answer correctness sensitive to minor errors in rationale. Therefore, we propose a robust post-thinking mechanism to generate answers before the rationale. Thanks to this answer-first setting: 1) the answer can escape from the rationale-sensitive problem; 2) the rationale serves as an error amplifier, making SLM focus on learning hard samples; and 3) the inferring efficiency can also benefit. Although post-thinking brings many advantages, it may lose the ability to analyze complex questions compared to pre-thinking. Therefore, a plug-and-play adaptive-thinking mechanism is proposed to integrate the merits of pre-thinking and post-thinking, in which a perception module based on soft prompt tuning is introduced to prompt SLM to answer or think first according to the complexity of questions. Extensive experiments are conducted across 12 datasets and 2 language models to demonstrate the effectiveness of the proposed mechanism.

Semantic segmentation of small-scale objects in very high resolution (VHR) remote sensing images plays an important role in some special tasks, such as change detection and mapping of land cover. However, due to small size, small-scale objects are more likely to be completely obscured by shadows than large-scale objects, which make it difficult for the traditional convolutional neural network (CNN) to distinguish small-scale objects from shadows. Furthermore, even if small-scale objects are distinguished, their boundaries are still difficult to refine. To solve the above problems, a novel context union edge network (CEN) for small-scale objects semantic segmentation is proposed by comprehensively considering both the contextual and edge information. In CEN, a plug-and-play context-based feature enhancement module (CFEM) is designed to enhance the ability of CNNs to distinguish small-scale objects. Then, an information exchange mechanism (IEM) is proposed based on the dual-stream (semantic and edge stream) network to refine the boundaries of small-scale objects. Finally, some experiments based on the ISPRS Vaihingen data set are conducted in terms of both overall accuracy (OA) and F1-score. The proposed CEN achieves 89.9% of F1-score for small-scale objects (cars) and 90.9% of OA, harvesting new state-of-the-art results.

Remarkable improvements have been seen in the semantic segmentation of remote-sensing images. As an effective structure to aggregate shallow information and deep information, encoder–decoder structure has been widely used in many state-of-the-art models, but it possesses two drawbacks that have not been fully addressed. On the one hand, encoder–decoder structure fuses the features obtained from shallow and deep layers directly; despite harvesting some detailed information, it also brings in noisy features owing to the poor discriminant ability of the shallow layers. On the other hand, existing encoder–decoder structure merely fuses the high-level information generated by the last layer of encoder once, which neglects its guidance ability to the feature aggregation process in the decoder. In this letter, we first propose an edge perception module (EPM) to eliminate the noisy features in the shallow information, as well as enhance features’ structural information. And then, we generate the most suitable guidance information adaptively for different stages in the decoder through high-level information module (HIM). Finally, we apply the guidance information to achieve feature aggregation in the feature aggregation module (FAM). Combined with EPM, HIM, and FAM, our proposed model achieves 89.5% overall accuracy (OA) on the challenging ISPRS Vaihingen test set, which is the new state-of-the-art in the semantic segmentation of remote-sensing images.

By reason of factors such as terrains, weather conditions, sensor imaging methods and cultural and economic development, there is a large shift between the remote sensing imagery collected from different geographic locations and different sensors, which makes the state-of-the-art semantic segmentation models trained on source domain (a image set gathered from specific geographic locations and sensors) difficult to generalize to target domain (another image set collected from other geographic locations and sensors). Currently, unsupervised domain adaptation using adversarial training whose purpose is to align the marginal distribution in the output space between source and target domain, is the most explored and practical approach to address this issue. However, this global alignment approach does not take into account diversities of different regions in a specific image nor the category-level distribution, which leads to the consequence that some regions and categories which are already well aligned between the source and target domain may be incorrectly remapped. Therefore, we propose a region and category adaptive domain discriminator, aiming to emphasize the differences in regions and categories during the process of alignment. Specifically, on the one hand, we propose an entropy-based regional attention module in domain discriminator to emphasize the importance of difficult-to-align regions. On the other hand, we propose a class-clear module to update only the distribution of existing categories in one iteration without affecting all categories. Finally, a lot of experiments are introduced to indicate that the proposed method can obtain better results when compared with other state-of-the-art unsupervised domain adaptation methods using adversarial training.

In this study, the authors explore a wireless network consisting of multiple smart terminals relying on multi‐mode in the presence of an eavesdropper, where the eavesdropper may be deployed to wiretap the wireless communications deliberately. In general, a smart terminal can be switched between the licensed modes (e.g. LTE) and the unlicensed modes (e.g. WiFi) according to user's demands (e.g. costs). The designed tradeoff is that the demands can be met by switching, but the security of wireless transmissions may be deteriorated, as the wireless communications can be overheard by the eavesdropper. The authors propose a transmission mode aware smart terminal scheduling (TM‐STS) scheme to improve the security of wireless transmissions of a smart terminal. Moreover, they analyse the security‐reliability tradeoff (SRT) of the TM‐STS scheme. They also present the SRT of the conventional round‐robin aided smart terminal scheduling (Rr‐STS) scheme. Numerical results show that the TM‐STS scheme outperforms the Rr‐STS scheme in terms of its SRT, and reveal that increasing the number of smart terminals not only can upgrade the SRT of the TM‐STS scheme, but also can be a major influence on the probability of mode switching between a licensed mode and an unlicensed mode of the TM‐STS scheme.

We study the physical-layer security of a cognitive radio system in the face of multiple eavesdroppers (EDs), which is composed of a secondary base station (SBS), multiple secondary users (SUs) as well as a pair of primary transmitter (PT) and primary receiver (PR), where the SUs first harvest energy from their received radio frequency signals transmitted by the PT and then communicate with the SBS relying on opportunistic scheduling. We consider two specific user scheduling schemes, namely, the channel-aware user scheduling (CaUS) and the energy-aware user scheduling (EaUS). In the CaUS scheme, an SU having the best instantaneous SU-SBS link (spanning from SUs to SBS) will be activated to communicate with the SBS. By contrast, the EaUS scheme takes into account both the amount of energy harvested from the PT and the instantaneous quality of the SU-SBS link. We analyze the security-reliability tradeoff (SRT) of both the CaUS and EaUS schemes in terms of their intercept versus outage probability. We also provide the SRT analysis of traditional round-robin user scheduling (RrUS) used as a benchmarker of the CaUS and EaUS schemes. We demonstrate that the EaUS scheme achieves the best outage and secrecy performance in the high main-to-eavesdropper ratio (MER) region, but a worse secrecy performance than the CaUS method in the low-MER region. Moreover, from a security versus reliability perspective, the CaUS outperforms both the EaUS and the RrUS in the low-MER region. Surprisingly, this also implies that although the user scheduling criterion of EaUS exploits the knowledge of both the amount of harvested power and instantaneous channel state information (CSI), it exhibits a degraded physical-layer security in the low-MER region due to the fact that the increased harvested energy is beneficial not only for the legitimate SBS receiver but also for the EDs.

This paper investigates secrecy energy efficiency (SEE) optimization problem for simultaneous wireless information and power transfer (SWIPT) in a multiple-input single-output (MISO) downlink system, in which the legitimate user (LU) exploits power splitting (PS) scheme for information decoding (ID) and energy harvesting (EH) simultaneously. In order to prevent eavesdroppers from disturbing the reception of the legitimate user, BS also emits artificial noise (AN). The objective is to maximize SEE by jointly optimizing the confidential signal power, AN power and PS ratio while guaranteeing the secrecy rate, maximum transmitted power and minimum harvested energy constraints. The problem is nonconvex and hard to solve, we first convert it into an easier subtractive form, and then propose an effective two-stage algorithm. Numerical results illustrate the efficiency of the algorithm and the benefits of AN in enhancing the SEE performance.

In this study, the authors investigate the physical‐layer security in a cognitive amplify‐and‐forward relay network consisting of a secondary transmitter (ST) and a secondary destination (SD) with the aid of multiple secondary relays (SRs) in the face of multiple eavesdroppers. In cognitive radio networks, increasing transmit power may not always be beneficial in terms of improving the channel capacity of cognitive transmissions, which would not only cause an extra interference to primary user, but also enhance the possibility of successfully intercepting the cognitive transmissions at an eavesdropper because an improved signal strength is received in this case. The authors propose two relay selection schemes to improve the physical‐layer security of cognitive transmissions against eavesdropping attacks, which are referred to as the global and partial channel state information based relay selection, denoted by GCSIbRS and PCSIbRS, respectively. The authors analyse the intercept probability of the proposed GCSIbRS and PCSIbRS, as well as the traditional round‐robin and all‐relay transmission schemes. It is shown that the proposed GCSIbRS and PCSIbRS schemes both outperform the conventional round‐robin and all‐relay schemes in terms of their intercept probability performance.

The ortholog conjecture posits that orthologous genes are functionally more similar than paralogous genes. This conjecture is a cornerstone of phylogenomics and is used daily by both computational and experimental biologists in predicting, interpreting, and understanding gene functions. A recent study, however, challenged the ortholog conjecture on the basis of experimentally derived Gene Ontology (GO) annotations and microarray gene expression data in human and mouse. It instead proposed that the functional similarity of homologous genes is primarily determined by the cellular context in which the genes act, explaining why a greater functional similarity of (within-species) paralogs than (between-species) orthologs was observed. Here we show that GO-based functional similarity between human and mouse orthologs, relative to that between paralogs, has been increasing in the last five years. Further, compared with paralogs, orthologs are less likely to be included in the same study, causing an underestimation in their functional similarity. A close examination of functional studies of homologs with identical protein sequences reveals experimental biases, annotation errors, and homology-based functional inferences that are labeled in GO as experimental. These problems and the temporary nature of the GO-based finding make the current GO inappropriate for testing the ortholog conjecture. RNA sequencing (RNA-Seq) is known to be superior to microarray for comparing the expressions of different genes or in different species. Our analysis of a large RNA-Seq dataset of multiple tissues from eight mammals and the chicken shows that the expression similarity between orthologs is significantly higher than that between within-species paralogs, supporting the ortholog conjecture and refuting the cellular context hypothesis for gene expression. We conclude that the ortholog conjecture remains largely valid to the extent that it has been tested, but further scrutiny using more and better functional data is needed.