Large Vision-Language Models (LVLMs) demonstrate impressive general-purpose capabilities but often suffer from catastrophic forgetting when incorporating specialized knowledge. To address this plasticity-stability dilemma, we introduce Structured Dialogue Fine-Tuning (SDFT), a data-centric approach that injects domain-specific concepts while preserving foundational abilities. Distinct from parameter-constrained continual learning methods, SDFT leverages a three-phase dialogue structure: Foundation Preservation reinforces pre-trained visual-linguistic alignment through captioning tasks; Contrastive Disambiguation uses carefully designed counterfactual examples to establish precise semantic boundaries; and Knowledge Specialization embeds specialized information via chain-of-thought reasoning. Evaluations across personalized entity recognition, abstract concept understanding, and biomedical domains show that SDFT significantly outperforms state-of-the-art baselines, including EWC-LoRA and O-LoRA. The results confirm SDFT’s effectiveness in balancing specialized knowledge acquisition and general capability retention.

Multimodal Aspect-based Sentiment Analysis (MABSA) aims to extract aspect-sentiment pairs from a combination of text and images. However, images often contain content that is either irrelevant to the textual information or not related to the sentiment prediction, which can adversely affect the accuracy of the model predictions. Furthermore, existing models neglect precise regional information beyond global image features, which could also assist in enhancing aspect-based sentiment prediction, but may also introduce noise that damages the model. To address these issues, this study proposes Knowledge-injected Mixture-of-Prefix (KMoP) to inject various types of knowledge into the language model and reduce external noise. Specifically, external knowledge is injected in the form of prefixes into the language model, which minimizes catastrophic forgetting issue and generate noise-insensitive representations. Additionally, to allow different layers of the language model to automatically select the required knowledge, we differentiate the aggregation of prefixes from different knowledge sources for each layer through Mixture-of-Prefix. This paper simultaneously divides the training process into two parts, with the first phase training on the original clean dataset and the second phase fine-tuning on the original dataset with added noise. KMoP achieves state-of-the-art performance on MABSA task, with extensive supplementary experiments demonstrating its enhanced robustness to noise.

In this article, the fuzzy dynamic event-triggered (DET) containment control problem for human-in-the-loop (HiTL) multiagent systems (MASs) with error constraints is investigated. Through utilization of fuzzy logic systems (FLSs), a high-gain state observer is presented to estimate unavailable states. To improve the transient performance, a fixed-time prescribed performance (FTPP) function is presented to restrict the containment errors and virtual errors. Under the backstepping control framework, a barrier Lyapunov function is designed to guarantee that the transform errors do not transgress the constraint bounds. Meanwhile, a nonlinear filter is introduced to obtain reduction calculation and enhance the control performance. Furthermore, the DET mechanism is presented to decrease the network communication burden. By directly controlling multiple leaders to form a dynamic convex hull, the proposed control strategy ensures that output signals of followers can converge to this convex hull and the containment errors can converge to the prescribed bounds within fixed time. Herein, a simulation example is presented to assess the effectiveness of the proposed control scheme.