Recent advances in large language models (LLMs) and text-aware graph learning have increased interest in reasoning over textattributed graphs (TAGs).In many real-world settings, such graphs are inherently heterogeneous, with most existing benchmarks remaining largely homogeneous in structure.As a result, the lack of large-scale benchmarks for heterogeneous text-attributed graphs has hindered systematic evaluation and fair comparison of existing methods.In this work, we introduce CITE -Catalytic Information Textual Entities Graph, the first and largest heterogeneous text-attributed citation graph benchmark for catalytic materials.CITE contains over 438K nodes and 1.2M edges spanning four node types and four relation types, with rich node-level textual information.We establish standardized evaluation protocols for node classification and link prediction, and conduct ablation studies to assess the impact of graph heterogeneity and textual attributes.Using CITE, we benchmark four classes of learning paradigms, including homogeneous graph models, heterogeneous graph models, LLM-centric models, and LLM+Graph models.By providing a largescale heterogeneous text-attributed benchmark together with standardized evaluation protocols and comprehensive baselines, CITE enables systematic assessment across diverse modeling paradigms and offers new insights into textaware and LLM-enhanced graph learning.The dataset 1 , codebase and evaluation suite 2 are publicly available.

In real-world scenarios, optimization problems generally exhibit multiobjective characteristics, necessitating a balance among conflicting evaluation criteria. An effective approach to multiobjective optimization (MOO) is to approximate the Pareto front. However, completely solving for the Pareto front incurs extremely high computational costs, making it almost infeasible in practical applications. In practical scenarios, decision makers (DMs) typically require only a single most preferred solution from the Pareto-optimal set. Existing optimization methods often struggle to incorporate real-time DMs' preferences during the optimization process. To address this limitation, this article proposes a preference-aware Bayesian optimization (PABO) framework for interactive decision-making that seamlessly integrates DMs' feedback throughout the entire optimization process. By embedding preference information into the candidate solution generation stage, PABO dynamically adjusts the balance between exploration of uncertain regions and exploitation of preference-aligned solutions, thereby achieving efficient preference satisfaction. Experiments on benchmark functions and real-world engineering cases demonstrate that PABO achieves comparable or superior solution quality with significantly fewer expensive evaluations than state-of-the-art methods. This achievement indicates that PABO demonstrates significant advantages in improving optimization efficiency and reducing costs, providing a more feasible technical approach for the practical application of MOO problems.

Continuous Relation Extraction (CRE) aims to incrementally learn relation knowledge from a non-stationary stream of data. Since the introduction of new relational tasks can overshadow previously learned information, catastrophic forgetting becomes a significant challenge in this domain. Current replay-based training paradigms prioritize all data uniformly and train memory samples through multiple rounds, which would result in overfitting old tasks and pronounced bias towards new tasks because of the imbalances of the replay set. To handle the problem, we introduce the DecouPled CRE (DP-CRE) framework that decouples the process of prior information preservation and new knowledge acquisition. This framework examines alterations in the embedding space as new relation classes emerge, distinctly managing the preservation and acquisition of knowledge. Extensive experiments show that DP-CRE significantly outperforms other CRE baselines across two datasets.

Considering that some small-scale algorithms do not require a model with large capacity, this paper designs a topology structure graph search framework based on differentiable search of basic mathematical elements, using gradient descent to optimize the connection relationship of the algorithm topology graph, and applies this basic operation-level search framework to the following three application scenarios: 1) equation search 2) image binary classification problem 3) reinforcement learning(RL) reward function search, all of which have achieved results. The validity and efficiency of this differentiable search method has been proven. Although the examples in all three scenarios are not very complicated, the significant decrease in losses in 2) and the significant in-crease in rewards in 3) are sufficient to demonstrate the validity of the algorithm, which provides inspiration for future applications such as reward search.

Reconstruction of a 12-lead electrocardiogram (ECG) from a serial 3-lead ECG has been researched in the past to satisfy the need for more wearing comfort and ambulatory situations. The accuracy and real-time performance of traditional methods need to be improved. In this study, we present a novel method based on convolutional neural networks (CNNs) for the synthesis of missing precordial leads. The results show that the proposed method receives better similarity and consumes less time using the PTB database. Particularly, the presented method shows outstanding performance in reconstructing the pathological ECG signal, which is crucial for cardiac diagnosis. Our CNN-based method is shown to be more accurate and time-saving for deployment in non-hospital situations to synthesize a standard 12-lead ECG from a reduced lead-set ECG recording. This is promising for real cardiac care.

The financial support information in the original article is incorrect. The correct information is given below: Project supported by the National Natural Science Foundation of China (No. 61702044), the Fundamental Research Funds for the Central Universities, China (No. 2017RC27), and the BUPT Excellent Ph.D. Students Foundation.