Hallucination remains a major challenge for Large Vision-Language Models (LVLMs). Direct Preference Optimization (DPO) has gained increasing attention as a simple solution to hallucination issues. It directly learns from constructed preference pairs that reflect the severity of hallucinations in responses to the same prompt and image. Nonetheless, different data construction methods in existing works bring notable performance variations. We identify a crucial factor here: outcomes are largely contingent on whether the constructed data aligns on-policy w.r.t the initial (reference) policy of DPO. Theoretical analysis suggests that learning from off-policy data is impeded by the presence of KL-divergence between the updated policy and the reference policy. From the perspective of dataset distribution, we systematically summarize the inherent flaws in existing algorithms that employ DPO to address hallucination issues. To alleviate the problems, we propose On-Policy Alignment (OPA)-DPO framework, which uniquely leverages expert feedback to correct hallucinated responses and aligns both the original and expert-revised responses in an on-policy manner. Notably, with only 4.8k data, OPADPO achieves an additional reduction in the hallucination rate of LLaVA-1.5-7B: 13.26% on the AMBER benchmark and 5.39% on the Object-Hal benchmark, compared to the previous SOTA algorithm trained with 16k samples.

Safe offline reinforcement learning (RL), which aims to learn the safety-guaranteed policy without risky online interaction with environments, has attracted growing recent attention for safety-critical scenarios. However, existing approaches encounter out-of-distribution problems during the testing phase, which can result in potentially unsafe outcomes. This issue arises due to the infinite possible combinations of reward-related and cost-related states. In this work, we propose State Decoupling with Q-supervised Contrastive representation (SDQC), a novel framework that decouples the global observations into reward- and cost-related representations for decision-making, thereby improving the generalization capability for unfamiliar global observations. Compared with the classical representation learning methods, which typically require model-based estimation (e.g., bisimulation), we theoretically prove that our Q-supervised method generates a coarser representation while preserving the optimal policy, resulting in improved generalization performance. Experiments on DSRL benchmark problems provide compelling evidence that SDQC surpasses other baseline algorithms, especially for its exceptional ability to achieve almost zero violations in more than half of the tasks, while the state-of-the-art algorithm can only achieve the same level of success in a quarter of the tasks. Further, we demonstrate that SDQC possesses superior generalization ability when confronted with unseen environments.

Real-world datasets collected from sensors or human inputs are prone to noise and errors, posing significant challenges for applying offline reinforcement learning (RL). While existing methods have made progress in addressing corrupted actions and rewards, they remain insufficient for handling corruption in high-dimensional state spaces and for cases where multiple elements in the dataset are corrupted simultaneously. Diffusion models, known for their strong denoising capabilities, offer a promising direction for this problem—but their tendency to overfit noisy samples limits their direct applicability. To overcome this, we propose **A**mbient **D**iffusion-**G**uided Dataset Recovery (**ADG**), a novel approach that pioneers the use of diffusion models to tackle data corruption in offline RL. First, we introduce Ambient Denoising Diffusion Probabilistic Models (DDPM) from approximated distributions, which enable learning on partially corrupted datasets with theoretical guarantees. Second, we use the noise-prediction property of Ambient DDPM to distinguish between clean and corrupted data, and then use the clean subset to train a standard DDPM. Third, we employ the trained standard DDPM to refine the previously identified corrupted data, enhancing data quality for subsequent offline RL training. A notable strength of ADG is its versatility—it can be seamlessly integrated with any offline RL algorithm. Experiments on a range of benchmarks, including MuJoCo, Kitchen, and Adroit, demonstrate that ADG effectively mitigates the impact of corrupted data and improves the robustness of offline RL under various noise settings, achieving state-of-the-art results.

This article explores the impact of source data compression on the performance of task execution at the receiver side of a communication system, and investigates the interference and impact of channel environment variations on semantic coding features. In order to further optimize the performance of the semantic communication model, a semantic communication framework (DIB-DeepSC) based on disentangled information bottleneck is proposed, which improves the inference accuracy of the model by separating and decoupling irrelevant information in the source data, thus compressing valid information related to downstream task execution to a greater extent, reduced communication overhead. Meanwhile, the influence of the Lagrange multiplier$\beta $in the classical information bottleneck (IB) framework is eliminated, which avoids the need to manually optimize$\beta $several times in the semantic communication model. And the dynamic coding method (DIB-DE) is further designed based on adaptive weights, which can dynamically adjust the coding features according to the channel conditions and enhance the robustness of the model. Numerous experiments show that the proposed DIB-DeepSC framework combined with the DIB-DE dynamic encoding communication scheme possesses better semantic extraction and task inference performance relative to the benchmark methods. This scheme is expected to realize more efficient and reliable semantic transmission in practical communication systems and provides new ideas for developing practical semantic communication systems.

Offline reinforcement learning (RL), which aims to fully explore offline datasets for training without interaction with environments, has attracted growing recent attention. A major challenge for the real-world application of offline RL stems from the robustness against state observation perturbations, e.g., as a result of sensor errors or adversarial attacks. Unlike online robust RL, agents cannot be adversarially trained in the offline setting. In this work, we propose Diffusion Model-Based Predictor (DMBP) in a new framework that recovers the actual states with conditional diffusion models for state-based RL tasks. To mitigate the error accumulation issue in model-based estimation resulting from the classical training of conventional diffusion models, we propose a non-Markovian training objective to minimize the sum entropy of denoised states in RL trajectory. Experiments on standard benchmark problems demonstrate that DMBP can significantly enhance the robustness of existing offline RL algorithms against different scales of ran- dom noises and adversarial attacks on state observations. Further, the proposed framework can effectively deal with incomplete state observations with random combinations of multiple unobserved dimensions in the test. Our implementation is available at https://github.com/zhyang2226/DMBP.

With the development of mobile edge computing and neural network Deep Learning (DL), more and more scholars are studying the combination of the two. This paper mainly studies the application of mobile edge computing and neural network DL in IoT computing collaboration and data-aware routing algorithms. Therefore, this paper proposes the deployment options of MEC technology and ETSIMEC in mobile edge computing, combining mobile edge computing with DL, designs an optimization algorithm based on Markov decision process and feature expression learning, and then analyzes and optimizes IoT computing and VANET routing algorithm. In order to have a clearer direction for the optimization algorithm, this paper also designs the edge computing model training and experiment comparison, the DL algorithm comparison experiment, and the routing algorithm simulation experiment and performance analysis. Combined with the experimental results, it is optimized and compared with traditional IoT computing and routing algorithms. Finally, it is concluded that the computing efficiency of IoT computing based on edge computing and DL designed in this paper is 21.33% higher than that of traditional IoT computing. The efficiency of the routing algorithm based on edge computing and DL designed in this paper is 9.29% higher than that of the traditional routing algorithm.

Named entity recognition (NER) is a key component of many natural language processing (NLP) applications. The majority of advanced research, however, has not been widely applied to low-resource languages represented by Malay due to the data-hungry problem. In this paper, we present a system for building a Malay NER dataset (MS-NER) of 20,146 sentences through labelled datasets of homologous languages and iterative optimisation. Additionally, we propose a Multi-Task framework, namely MTBR, to integrate boundary information more effectively for NER. Specifically, boundary detection is treated as an auxiliary task and an enhanced Bidirectional Revision module with a gated ignoring mechanism is proposed to undertake conditional label transfer. This can reduce error propagation by the auxiliary task. We conduct extensive experiments on Malay, Indonesian, and English. Experimental results show that MTBR could achieve competitive performance and tends to outperform multiple baselines. The constructed dataset and model would be made available to the public as a new, reliable benchmark for Malay NER.

To solve the problems of incomplete topic description and repetitive crawling of visited hyperlinks in traditional focused crawling methods, in this paper, we propose a novel focused crawler using an improved tabu search algorithm with domain ontology and host information (FCITS_OH), where a domain ontology is constructed by formal concept analysis to describe topics at the semantic and knowledge levels. To avoid crawling visited hyperlinks and expand the search range, we present an improved tabu search (ITS) algorithm and the strategy of host information memory. In addition, a comprehensive priority evaluation method based on Web text and link structure is designed to improve the assessment of topic relevance for unvisited hyperlinks. Experimental results on both tourism and rainstorm disaster domains show that the proposed focused crawlers overmatch the traditional focused crawlers for different performance metrics.

High-speed railway (HSR) wireless communications are required to ensure strict security. In this work, we study the secrecy performance of a nonorthogonal multiple access- (NOMA-) aided HSR wireless communication system in the case of an eavesdropping user. Specifically, applying NOMA technology to the HSR communication system can effectively improve the data rates. Therefore, we study the secrecy performance of the downlink NOMA system under the HSR wireless communication. In particular, the exact analytical results for the secrecy outage probability (SOP) based on no small-scale channel state information (CSI) are derived. We also provide all of the parameterizations for the proposed channel model. Finally, the correctness of theoretical derivation is verified via simulations. Results show the positive effect of utilizing the NOMA for enhancing wireless systems secrecy performance.