The powerful text understanding and generation capabilities of large language models (LLMs) have brought new vitality to general recommendation with implicit feedback. One possible strategy involves generating a unique user (or item) profile from historical interaction data, which is then mapped to a semantic representation in the language space. However, a single-instance profile may be insufficient to comprehensively capture the complex intentions behind a user's interacted items. Moreover, due to the inherent instability of LLMs, a biased or misinterpreted profile could even undermine the original recommendation performance. Consequently, an intuitive solution is to generate multiple profiles for each user (or item), each reflecting a distinct aspect of their characteristics. In light of this, we propose a unified recommendation framework with multi-faceted profile extrapolation (ProEx) in this paper. By leveraging chain-of-thought reasoning, we construct multiple distinct profiles for each user and item. These new profiles are subsequently mapped into semantic vectors, extrapolating from the position of the original profile to explore a broader region of the language space. Subsequently, we introduce the concept of environments, where each environment represents a possible linear combination of all profiles. The differences across environments are minimized to reveal the inherent invariance of user preferences. We apply ProEx to three discriminative methods and three generative methods, and conduct extensive experiments on three datasets. The experimental results demonstrate that ProEx significantly enhances the performance of these base recommendation models.

Generative recommendation aims to learn the underlying generative process over the entire item set to produce recommendations for users. Although it leverages non-linear probabilistic models to surpass the limited modeling capacity of linear factor models, it is often constrained by a trade-off between representation ability and tractability. With the rise of a new generation of generative methods based on pre-trained language models (LMs), incorporating LMs into general recommendation with implicit feedback has gained considerable attention. However, adapting them to generative recommendation remains challenging. The core reason lies in the mismatch between the input-output formats and semantics of generative models and LMs, making it challenging to achieve optimal alignment in the feature space. This work addresses this issue by proposing a model-agnostic generative recommendation framework called DMRec, which introduces a probabilistic meta-network to bridge the outputs of LMs with user interactions, thereby enabling an equivalent probabilistic modeling process. Subsequently, we design three cross-space distribution matching processes aimed at maximizing shared information while preserving the unique semantics of each space and filtering out irrelevant information. We apply DMRec to three different types of generative recommendation methods and conduct extensive experiments on three public datasets. The experimental results demonstrate that DMRec can effectively enhance the recommendation performance of these generative models, and it shows significant advantages over mainstream LM-enhanced recommendation methods.

Intent-based recommender systems have garnered significant attention for uncovering latent fine-grained preferences.Intents, as underlying factors of interactions, are crucial for improving recommendation interpretability.Most methods define intents as learnable parameters updated alongside interactions.However, existing frameworks often overlook textual information (e.g., user reviews, item descriptions), which is crucial for alleviating the sparsity of interaction intents.Exploring these multimodal intents, especially the inherent differences in representation spaces, poses two key challenges: i) How to align multimodal intents and effectively mitigate noise issues; ii) How to extract and match latent key intents across modalities.To tackle these challenges, we propose a modelagnostic framework, Intent Representation Learning with Large Language Model (IRLLRec), which leverages large language models (LLMs) to construct multimodal intents and enhance recommendations.Specifically, IRLLRec employs a dual-tower architecture to learn multimodal intent representations.Next, we propose pairwise and translation alignment to eliminate inter-modal differences and enhance robustness against noisy input features.Finally, to better match textual and interaction-based intents, we employ momentum distillation to perform teacher-student learning on fused intent representations.Empirical evaluations on three datasets show that our IRLLRec framework outperforms baselines 1 .

Contrastive Learning (CL)-based recommender systems have gained prominence in the context of Heterogeneous Graph (HG) due to their capacity to enhance the consistency of representations across different views. However, existing frameworks often neglect the fact that user-item interactions within HG are governed by diverse latent intents (e.g., brand preferences or demographic characteristics of item audiences), which are pivotal in capturing fine-grained relations. The exploration of these underlying intents, particularly through the lens of meta-paths in HGs, presents us with two principal challenges: i) How to integrate CL with intents; ii) How to mitigate noise from meta-path-driven intents. To address these challenges, we propose an innovative framework termedIntent-guided Heterogeneous Graph Contrastive Learning(IHGCL), which designed to enhance CL-based recommendation by capturing the intents contained within meta-paths. Specifically, the IHGCL framework includes: i) a meta-path-based Dual Contrastive Learning (DCL) approach to effectively integrate intents into the recommendation, constructing intent-intent contrast and intent-interaction contrast; ii) a Bottlenecked AutoEncoder (BAE) that combines mask propagation with the information bottleneck principle to significantly reduce noise perturbations introduced by meta-paths. Empirical evaluations conducted across six distinct datasets demonstrate the superior performance of our IHGCL framework relative to conventional baseline methods.

In recommender systems, heterogeneous graph neural networks (HGNNs) have demonstrated remarkable efficacy due to their capacity to harness rich auxiliary information within heterogeneous information networks (HINs). However, existing HGNN-based recommendation faces severe noise cascading challenge. The presence of substantial data noise can adversely affect robustness of recommender, as the graph structures are susceptible to noise and even unnoticed malicious perturbations. Moreover, these noises can propagate and accumulate through connected nodes, potentially exerting a profound impact on target nodes within the graph structure. To tackle the noise challenges, we present a Bottlenecked Heterogeneous Graph Contrastive Learning (BHGCL), aiming to enhance the robustness of recommendation systems. BHGCL can first effectively separate fine-grained latent factors from complex self-supervision signals with a disentangled-based encoder, leveraging diverse semantic information across various meta-paths. Then, by employing the information bottleneck (IB) principle, BHGCL adaptively learns to reduce noise in augmented graphs. IB can capture the minimum sufficient information from the data features, which significantly improves system performance in environments with noisy data. Experimental findings from multiple real-world datasets reveal that our approach surpasses the latest advanced recommendation systems, verifying its effectiveness and robustness. To reproduce our work, we have open-sourced our code at https://github.com/DuellingSword/BHGCL .

Heterogeneous graph neural networks (HGNN) have exhibited significant performance gains by modeling the information propagation process in graph-structured data for recommender systems. However, existing HGNN-based Recommendation still face two challenges: (1) They overlook the rich semantics brought by the combination of different meta-paths, making it difficult to capture the importance of various meta-paths; (2) when HGNN use meta-paths to capture high-order information, they are susceptible to noise data, as noise from connected nodes can create cumulative effects on a target node in the graph. To tackle these issues, we propose a new model called the Denoising Heterogeneous Graph Pre-training Framework (DHGPF) to enhance recommendation tasks. This framework has two stages: pre-training and training. In the pre-training stage, we assign learnable weights to different meta-paths and use a simplified multi-layer graph convolution network to automatically aggregate semantic information from different meta-path combinations. This approach can capture the importance of these paths. The training stage focuses on reducing noise using gating mechanism and denoising structure learning methods. These methods accomplish the denoising process through information filtering. Our model was evaluated on three real-world datasets, demonstrating that DHGPF outperforms other state-of-the-art recommendation methods. We have further organized the source code of the article at https://github.com/wangyu0627/DHGPF .