Existing LLM-driven recommendation systems (RS) suffer from over-reliance on a single pre-trained model, which limits adaptability across diverse scenarios due to differences in large language models' strengths in semantics, knowledge, and reasoning. To address this issue, we propose AMLrec (Adaptive Multi-LLM Recommendation), a dynamic routing-based adaptive multi-LLM collaboration framework that unifies two dominant paradigms—LLM as Recommender and LLM + Recommender—through decision knowledge complementation. For each user or item, a lightweight encoder generates embeddings that are compared with learnable LLM prototypes using cosine similarity to select the most suitable models. In the first paradigm, selected LLMs generate recommendations via structured prompts, and their outputs are aggregated to form the final recommendation list. In the second paradigm, chosen LLMs produce semantic embeddings, which are fused with learnable embeddings after PCA-based dimensionality reduction and aligned using a lightweight adapter to bridge distribution gaps. Notably, AMLrec does not require fine-tuning of the underlying LLMs, significantly reducing computational overhead. Experiments on real-world datasets demonstrate that the proposed approach consistently outperforms single-LLM baselines across all evaluation metrics, validating its effectiveness. The main contributions of this work are threefold: introducing dynamic routing for multi-LLM recommendation system collaboration, proposing a unified architecture that harmonizes both paradigms, and enabling efficient adaptation without LLM fine-tuning. The code is available at https://github.com/Jiale-12138/AMLrec.

Multi-behavior recommender systems improve prediction accuracy of target behaviors (e.g., purchases) by integrating auxiliary behaviors (e.g., page views). However, existing models face two key limitations: (1) Static propagation mechanisms and inflexible dependency modeling fail to capture dynamic changes in user preferences and cascading relationships between behaviors; (2) Sparse target behavior data usually leads to excessive influence of auxiliary signals, which degrades recommendation quality. To address these challenges, we propose the Dynamic Dependency-Aware Collaborative Contrastive Learning Multi-Behavior Recommendation Model, MBDCC. MBDCC has two dedicated modules: (1) Behavioral gating cascade and cross-attention fusion module, which dynamically models cascading dependencies between behaviors through learnable gate control transfer units controlled by behavioral attributes. This replaces static propagation with adaptive feature flow regulation, capturing evolving user preferences. Meanwhile, it uses a target-guided cross-attention mechanism to selectively fuse semantically relevant auxiliary signals using the target behavior as a query, addressing inflexible cross-behavioral dependency modeling; (2) Collaborative semantic enhancement module, it constructs a user similarity measure matrix based on co-occurrence frequency of interaction items in target behavior, and clusters nodes using a hybrid clustering strategy. By introducing contrastive learning between nodes and their clustering centers, the collaborative semantic information between similar nodes under the target behavior is effectively captured and amplified, alleviating the challenge of sparse target interaction data. Extensive experiments on three real-world datasets show that MBDCC consistently outperforms state-of-the-art baselines, it achieves remarkable improvements of up to 6.84% in Recall@10 and 5.18% in NDCG@50. Moreover, ablation experiments further demonstrate the correctness of our motivation and the necessity of the various modules of the MBDCC model.

Convolutional neural networks (CNNs) have been successfully applied to music genre classification tasks. With the development of diverse music, genre fusion has become common. Fused music exhibits multiple similar musical features such as rhythm, timbre, and structure, which typically arise from the temporal information in the spectrum. However, traditional CNNs cannot effectively capture temporal information, leading to difficulties in distinguishing fused music. To address this issue, this study proposes a CNN model called MusicNeXt for music genre classification. Its goal is to enhance the feature extraction method to increase focus on musical features, and increase the distinctiveness between different genres, thereby reducing classification result bias. Specifically, we construct the feature extraction module which can fully utilize temporal information, thereby enhancing its focus on music features. It exhibits an improved understanding of the complexity of fused music. Additionally, we introduce a genre-sensitive adjustment layer that strengthens the learning of differences between different genres through within-class angle constraints. This leads to increased distinctiveness between genres and provides interpretability for the classification results. Experimental results demonstrate that our proposed MusicNeXt model outperforms baseline networks and other state-of-the-art methods in music genre classification tasks, without generating category bias in the classification results.

Click-through rate prediction plays a critical role in many fields, and many efforts are devoted to analyzing item behavior as a way to find the interest preferences of target user. Most previous work has not taken into account and fully explored the basic information, historical behavior and temporal information of similar users, which can result in missing information and lead to inaccurate CTR prediction. To solve the above problem, we propose Item Multi-Information Evolution Network (IMIEN). Firstly, for the basic information of users, we use RNN sequences to capture the evolutionary dynamics of users interested in the target item. Secondly, we use Evolutionary Interest Extraction Block (EIEB) to mine the evolutionary interest of target user and similar users over time. Finally, we introduce temporal information to find recent user groups with similar interests as the target user, which aids in predicting the probability of the target user clicking on the target item. We achieve the best results on five public datasets compared with previous mainstream models, which validates the effectiveness of IMIEN.

Abstract Recently, many deep learning-based models have been successfully applied to click-through rate prediction. However, most previous models focus only on feature-level interactions between a single user behavior and the target item or only treat the user’s historical behavior as a sequence to uncover the hidden interests behind it when mining user interests. This can lead to user interest that evolves over time dynamically being ignored or the interest shown by a single user’s behavior not being exploited. Based on the above problems, we propose evolving interest with feature co-action network (EIFCN). Specifically, we first design user dynamic interest network to treat the user’s historical behavior as a sequence of information, and tap into the user’s hidden interests over time. In this part, we use a multi-head self-attention mechanism to initially process the data and then pass it into the deep learning network. Then a feature co-action network is designed to mine the user’s single behavior and the displayed feature-level interactions of the target item. Experimental results show that the EIFCN model performs better than other models.