Self-training (ST) has come to fruition in language understanding tasks by producing pseudo labels, which reduces the labeling bottleneck of language model fine-tuning. Nevertheless, in facilitating semi-supervised controllable language generation, ST faces two key challenges. First, augmented by self-generated pseudo text, generation models tend to over-exploit the previously learned text distribution, suffering from mode collapse and poor generation diversity. Second, generating pseudo text in each iteration is time-consuming, severely decelerating the training process. In this work, we propose KEST, a novel and efficient self-training framework to handle these problems. KEST utilizes a kernel-based loss, rather than standard cross entropy, to learn from the soft pseudo text produced by a shared non-autoregressive generator. We demonstrate both theoretically and empirically that KEST can benefit from more diverse pseudo text in an efficient manner, which allows not only refining and exploiting the previously fitted distribution but also enhanced exploration towards a larger potential text space, providing a guarantee of improved performance. Experiments on three controllable generation tasks demonstrate that KEST significantly improves control accuracy while maintaining comparable text fluency and generation diversity against several strong baselines.

The integration of renewable energy sources (RES) into modern power systems has become increasingly important due to climate change and macroeconomic and geopolitical instability. Among the RES, photovoltaic (PV) energy is rapidly emerging as one of the world's most promising. However, its widespread adoption poses challenges related to its inherently uncertain nature that can lead to imbalances in the electrical system. Therefore, accurate forecasting of PV production can help resolve these uncertainties and facilitate the integration of PV into modern power systems. Currently, PV forecasting methods can be divided into two main categories: physics-based and data-based strategies, with AI-based models providing state-of-the-art performance in PV power forecasting. However, while these AI-based models can capture complex patterns and relationships in the data, they ignore the underlying physical prior knowledge of the phenomenon. Therefore, we propose MATNet, a novel self-attention transformer-based architecture for multivariate multi-step day-ahead PV power generation forecasting. It consists of a hybrid approach that combines the AI paradigm with the prior physical knowledge of PV power generation of physics-based methods. The model is fed with historical PV data and historical and forecast weather data through a multi-level joint fusion approach. The effectiveness of the proposed model is evaluated using the Ausgrid benchmark dataset with different regression performance metrics. The results show that our proposed architecture significantly outperforms the current state-of-the-art methods with an RMSE equal to 0.0460. These findings demonstrate the potential of MATNet in improving forecasting accuracy and suggest that it could be a promising solution to facilitate the integration of PV energy into the power grid.

Federated Graph Neural Network (FedGNN) has recently emerged as a rapidly growing research topic, as it integrates the strengths of graph neural networks and federated learning to enable advanced machine learning applications without direct access to sensitive data. Despite its advantages, the distributed nature of FedGNN introduces additional vulnerabilities, particularly backdoor attacks stemming from malicious participants. Although graph backdoor attacks have been explored, the compounded complexity introduced by the combination of GNNs and federated learning has hindered a comprehensive understanding of these attacks, as existing research lacks extensive benchmark coverage and in-depth analysis of critical factors. To address these limitations, we propose Bkd-FedGNN, a benchmark for backdoor attacks on FedGNN. Specifically, Bkd-FedGNN decomposes the graph backdoor attack into trigger generation and injection steps, and extending the attack to the node-level federated setting, resulting in a unified framework that covers both node-level and graph-level classification tasks. Moreover, we thoroughly investigate the impact of multiple critical factors in backdoor attacks on FedGNN. These factors are categorized into global-level and local-level factors, including data distribution, the number of malicious attackers, attack time, overlapping rate, trigger size, trigger type, trigger position, and poisoning rate. Finally, we conduct comprehensive evaluations on 13 benchmark datasets and 13 critical factors, comprising 1,725 experimental configurations for node-level and graph-level tasks from six domains. These experiments encompass over 8,000 individual tests, allowing us to provide a thorough evaluation and insightful observations that advance our understanding of backdoor attacks on FedGNN.The Bkd-FedGNN benchmark is publicly available at https://github.com/usail-hkust/BkdFedGCN.

Multi-modal knowledge graph (MKG) includes triplets that consist of entities and relations and multi-modal auxiliary data. In recent years, multi-hop multi-modal knowledge graph reasoning (MMKGR) based on reinforcement learning (RL) has received extensive attention because it addresses the intrinsic incompleteness of MKG in an interpretable manner. However, its performance is limited by empirically designed rewards and sparse relations. In addition, this method has been designed for the transductive setting where test entities have been seen during training, and it works poorly in the inductive setting where test entities do not appear in the training set. To overcome these issues, we propose TMR (Topology-aware Multi-hop Reasoning), which can conduct MKG reasoning under inductive and transductive settings. Specifically, TMR mainly consists of two components. (1) The topology-aware inductive representation captures information from the directed relations of unseen entities, and aggregates query-related topology features in an attentive manner to generate the fine-grained entity-independent features. (2) After completing multi-modal feature fusion, the relation-augment adaptive RL conducts multi-hop reasoning by eliminating manual rewards and dynamically adding actions. Finally, we construct new MKG datasets with different scales for inductive reasoning evaluation. Experimental results demonstrate that TMP outperforms state-of-the-art MKGR methods under both inductive and transductive settings.

Existing controllable dialogue generation work focuses on the single-attribute control and lacks generalization capability to out-of-distribution multiple attribute combinations. In this paper, we explore the compositional generalization for multi-attribute controllable dialogue generation where a model can learn from seen attribute values and generalize to unseen combinations. We propose a prompt-based disentangled controllable dialogue generation model, DCG. It learns attribute concept composition by generating attribute-oriented prompt vectors and uses a disentanglement loss to disentangle different attributes for better generalization. Besides, we design a unified reference-free evaluation framework for multiple attributes with different levels of granularities. Experiment results on two benchmarks prove the effectiveness of our method and the evaluation metric.

However, due to the limitation of available resources, it is impractical to store all facts in KGs, which leads to the incompleteness [5], and the algorithms of KGC are required to solve the problem. There are a lot of researches focusing on KGC or link prediction tasks aiming to infer missing facts automatically based on known facts. Pioneering additive models [6-8] take the transformation from head entities to tail entities as a translation problem, while multiplicative models [9-12] try to measure the plausibility of unknown triplets by applying proper semantic similarity-based score function. Benefiting from the development of neural networks, several works concentrate on the deeper nonlinear interactions among entities and relations with innovative model structures [13-18]. Furthermore, some recent studies introduce GCN to take the structure information into consideration by aggregating neighborhood information [19-22], which brings significant improvement. Despite the high-performance of them, they fail to utilize direction information implied in different neighbors while merging them, which is important for making reasonable predictions. As shown in Figure 1, there exists original and inverse edges (relations) in KGs, according to the direction of them, the link prediction tasks can be divided into forward and backward sub-tasks, neighbors can also be grouped into forward and backward neighbors, and sub-tasks in different directions always have diverse preferences for neighbors. For example, while dealing with forward sub-task (Stan Lee, profession,?), it can be solved from backward neighbors including Spider-Man, Iron Man and Captain America; on the other hand, the answer for query (Stan Lee, ethnicity

Knowledge graphs (KGs) store rich facts about the real world. In this paper, we study KG alignment, which aims to find alignment between not only entities but also relations and classes in different KGs. Alignment at the entity level can cross-fertilize alignment at the schema level. We propose a new KG alignment approach, called DAAKG, based on deep learning and active learning. With deep learning, it learns the embeddings of entities, relations and classes, and jointly aligns them in a semi-supervised manner. With active learning, it estimates how likely an entity, relation or class pair can be inferred, and selects the best batch for human labeling. We design two approximation algorithms for efficient solution to batch selection. Our experiments on benchmark datasets show the superior accuracy and generalization of DAAKG and validate the effectiveness of all its modules.

Learning deep discrete latent presentations offers a promise of better symbolic and summarized abstractions that are more useful to subsequent downstream tasks. Inspired by the seminal Vector Quantized Variational Auto-Encoder (VQ-VAE), most of work in learning deep discrete representations has mainly focused on improving the original VQ-VAE form and none of them has studied learning deep discrete representations from the generative viewpoint. In this work, we study learning deep discrete representations from the generative viewpoint. Specifically, we endow discrete distributions over sequences of codewords and learn a deterministic decoder that transports the distribution over the sequences of codewords to the data distribution via minimizing a WS distance between them. We develop further theories to connect it with the clustering viewpoint of WS distance, allowing us to have a better and more controllable clustering solution. Finally, we empirically evaluate our method on several well-known benchmarks, where it achieves better qualitative and quantitative performances than the other VQ-VAE variants in terms of the codebook utilization and image reconstruction/generation.

Large Language Models (LLMs) have demonstrated impressive capabilities in generating fluent text, as well as tendencies to reproduce undesirable social biases. This study investigates whether LLMs reproduce the moral biases associated with political groups in the United States, an instance of a broader capability herein termed moral mimicry. This hypothesis is explored in the GPT-3/3.5 and OPT families of Transformer-based LLMs. Using tools from Moral Foundations Theory, it is shown that these LLMs are indeed moral mimics. When prompted with a liberal or conservative political identity, the models generate text reflecting corresponding moral biases. This study also explores the relationship between moral mimicry and model size, and similarity between human and LLM moral word use.

This disease, characterized by cognitive impairments such as memory loss, predominantly affects aging populations, exerting an escalating burden on global healthcare systems as societies continue to age [3]. The significance of AD is further magnified by the increasing life expectancy globally, with the disease now recognized as a leading cause of disability and dependency among older people [4]. Consequently, AD has substantial social, economic, and health system implications, making its understanding and awareness of paramount importance [5, 6]. Despite the ubiquity and severity of AD, a gap persists in comprehensive, data-driven public understanding of this complex health narrative. Traditionally, public health professionals have to rely on labor-intensive methods such as web scraping, API data collection, data postprocessing, and analysis/synthesis to gather insights from news media, health reports, and other textual sources [7, 8, 9].