ACM Transactions on Information Systems最新文献

In recent years, various dense retrieval methods have been developed to improve the performance of search engines with a vectorized index. However, these approaches require a large pre-computed index and have limited capacity to memorize all semantics in a document within a single vector. To address these issues, researchers have explored end-to-end generative retrieval models that use a seq-to-seq generative model to directly return identifiers of relevant documents. Although these models have been effective, they are often trained with the maximum likelihood estimation method. It only encourages the model to assign a high probability to the relevant document identifier, ignoring the relevance comparisons of other documents. This may lead to performance degradation in ranking tasks, where the core is to compare the relevance between documents. To address this issue, we propose a ranking-oriented generative retrieval model that incorporates relevance signals in order to better estimate the relative relevance of different documents in ranking tasks. Based upon the analysis of the optimization objectives of dense retrieval and generative retrieval, we propose utilizing dense retrieval to provide relevance feedback for generative retrieval. Under an alternate training framework, the generative retrieval model gradually acquires higher-quality ranking signals to optimize the model. Experimental results show that our approach increasing Recall@1 by 12.9% with respect to the baselines on MS MARCO dataset.

Visually-aware recommender systems have found widespread applications in domains where visual elements significantly contribute to the inference of users’ potential preferences. While the incorporation of visual information holds the promise of enhancing recommendation accuracy and alleviating the cold-start problem, it is essential to point out that the inclusion of item images may introduce substantial security challenges. Some existing works have shown that the item provider can manipulate item exposure rates to its advantage by constructing adversarial images. However, these works cannot reveal the real vulnerability of visually-aware recommender systems because (1) the generated adversarial images are markedly distorted, rendering them easily detected by human observers; (2) the effectiveness of these attacks is inconsistent and even ineffective in some scenarios or datasets. To shed light on the real vulnerabilities of visually-aware recommender systems when confronted with adversarial images, this paper introduces a novel attack method, IPDGI (Item Promotion by Diffusion Generated Image). Specifically, IPDGI employs a guided diffusion model to generate adversarial samples designed to promote the exposure rates of target items (e.g., long-tail items). Taking advantage of accurately modeling benign images’ distribution by diffusion models, the generated adversarial images have high fidelity with original images, ensuring the stealth of our IPDGI. To demonstrate the effectiveness of our proposed methods, we conduct extensive experiments on two commonly used e-commerce recommendation datasets (Amazon Beauty and Amazon Baby) with several typical visually-aware recommender systems. The experimental results show that our attack method significantly improves both the performance of promoting the long-tailed (i.e., unpopular) items and the quality of generated adversarial images.

Maximum inner product search (MIPS) over dense and sparse vectors have progressed independently in a bifurcated literature for decades; the latter is better known as top-(k) retrieval in Information Retrieval. This duality exists because sparse and dense vectors serve different end goals. That is despite the fact that they are manifestations of the same mathematical problem. In this work, we ask if algorithms for dense vectors could be applied effectively to sparse vectors, particularly those that violate the assumptions underlying top-(k) retrieval methods. We study clustering-based approximate MIPS where vectors are partitioned into clusters and only a fraction of clusters are searched during retrieval. We conduct a comprehensive analysis of dimensionality reduction for sparse vectors, and examine standard and spherical KMeans for partitioning. Our experiments demonstrate that clustering-based retrieval serves as an efficient solution for sparse MIPS. As byproducts, we identify two research opportunities and explore their potential. First, we cast the clustering-based paradigm as dynamic pruning and turn that insight into a novel organization of the inverted index for approximate MIPS over general sparse vectors. Second, we offer a unified regime for MIPS over vectors that have dense and sparse subspaces, that is robust to query distributions.

Existing sequential POI recommendation methods overlook a fact that each city exhibits distinct characteristics and totally ignore the city signature. In this study, we claim that city matters in sequential POI recommendation and fully exploring city signature can highlight the characteristics of each city and facilitate cross-city complementary learning. To this end, we consider the two-city scenario and propose a dual-target cross-city sequential POI recommendation model (DCSPR) to achieve the purpose of complementary learning across cities. On one hand, DCSPR respectively captures geographical and cultural characteristics for each city by mining intra-city regions and intra-city functions of POIs. On the other hand, DCSPR builds a transfer channel between cities based on intra-city functions, and adopts a novel transfer strategy to transfer useful cultural characteristics across cities by mining inter-city functions of POIs. Moreover, to utilize these captured characteristics for sequential POI recommendation, DCSPR involves a new region- and function-aware network for each city to learn transition patterns from multiple views. Extensive experiments conducted on two real-world datasets with four cities demonstrate the effectiveness of DCSPR.

Providing potential next point-of-interest (POI) suggestions for users has become a prominent task in location-based social networks, which receives more and more attention from the industry and academia, and it remains challenging due to highly dynamic and personalized interactions in user movements. Currently, state-of-the-art works develop various graph- and sequential-based learning methods to model user-POI interactions and transition regularities. However, there are still two significant shortcomings in these works: (1) Ignoring personalized spatial- and temporal-aspect interactive characteristics capable of exhibiting periodic interests of users; (2) Insufficiently leveraging the sequential patterns of interactions for beyond-pairwise high-order collaborative signals among users’ sequences. To jointly address these challenges, we propose a novel multi-view hypergraph learning with spatial-temporal periodic interests for next POI recommendation (MvStHgL). In the local view, we attempt to learn the POI representation of each interaction via jointing periodic characteristics of spatial and temporal aspects. In the global view, we design a hypergraph by regarding interactive sequences as hyperedges to capture high-order collaborative signals across users, for further POI representations. More specifically, the output of POI representations in the local view is used for the initialized embedding, and the aggregation and propagation in the hypergraph are performed by a novel Node-to-Hypergraph-to-Node scheme. Furthermore, the captured POI embeddings are applied to achieve sequential dependency modeling for next POI prediction. Extensive experiments on three real-world datasets demonstrate that our proposed model outperforms the state-of-the-art models.

A Session-Based Recommendation (SBR) seeks to predict users’ future item preferences by analyzing their interactions with previously clicked items. In recent approaches, Graph Neural Networks (GNNs) have been commonly applied to capture item relations within a session to infer user intentions. However, these GNN-based methods typically struggle with feature ambiguity between the sequential session information and the item conversion within an item graph, which may impede the model's ability to accurately infer user intentions. In this paper, we propose a novel Multi-hop Multi-view Memory Transformer ((rm{M^{3}T})) to effectively integrate the sequence-view information and relation conversion (graph-view information) of items in a session. First, we propose a Multi-view Memory Transformer ((rm{M^{2}T})) module to concurrently obtain multi-view information of items. Then, a set of trainable memory matrices are employed to store sharable item features, which mitigates cross-view item feature ambiguity. To comprehensively capture latent user intentions, a Multi-hop (rm{M^{2}T}) ((rm{M^{3}T})) framework is designed to integrate user intentions across different hops of an item graph. Specifically, a k-order power method is proposed to manage the item graph to alleviate the over-smoothing problem when obtaining high-order relations of items. Extensive experiments conducted on three real-world datasets demonstrate the superiority of our method.

Unleashing the power of image-text matching in real-world applications is hampered by noisy correspondence. Manually curating high-quality datasets is expensive and time-consuming, and datasets generated using diffusion models are not adequately well-aligned. The most promising way is to collect image-text pairs from the Internet, but it will inevitably introduce noisy correspondence. To reduce the negative impact of noisy correspondence, we propose a novel model that first transforms the noisy correspondence filtering problem into a similarity distribution modeling problem by exploiting the powerful capabilities of pre-trained models. Specifically, we use the Gaussian Mixture model to model the similarity obtained by CLIP as clean distribution and noisy distribution, to filter out most of the noisy correspondence in the dataset. Afterward, we used relatively clean data to fine-tune the model. To further reduce the negative impact of unfiltered noisy correspondence, i.e., a minimal part where two distributions intersect during the fine-tuning process, we propose a distribution-sensitive dynamic margin ranking loss, further increasing the distance between the two distributions. Through continuous iteration, the noisy correspondence gradually decreases and the model performance gradually improves. Our extensive experiments demonstrate the effectiveness and robustness of our model even under high noise rates.

Reinforcement learning serves as a potent tool for modeling dynamic user interests within recommender systems, garnering increasing research attention of late. However, a significant drawback persists: its poor data efficiency, stemming from its interactive nature. The training of reinforcement learning-based recommender systems demands expensive online interactions to amass adequate trajectories, essential for agents to learn user preferences. This inefficiency renders reinforcement learning-based recommender systems a formidable undertaking, necessitating the exploration of potential solutions. Recent strides in offline reinforcement learning present a new perspective. Offline reinforcement learning empowers agents to glean insights from offline datasets and deploy learned policies in online settings. Given that recommender systems possess extensive offline datasets, the framework of offline reinforcement learning aligns seamlessly. Despite being a burgeoning field, works centered on recommender systems utilizing offline reinforcement learning remain limited. This survey aims to introduce and delve into offline reinforcement learning within recommender systems, offering an inclusive review of existing literature in this domain. Furthermore, we strive to underscore prevalent challenges, opportunities, and future pathways, poised to propel research in this evolving field.

Recently, the user-side fairness issue in Collaborative Filtering (CF) algorithms has gained considerable attention, arguing that results should not discriminate an individual or a sub user group based on users’ sensitive attributes (e.g., gender). Researchers have proposed fairness-aware CF models by decreasing statistical associations between predictions and sensitive attributes. A more natural idea is to achieve model fairness from a causal perspective. The remaining challenge is that we have no access to interventions, i.e., the counterfactual world that produces recommendations when each user have changed the sensitive attribute value. To this end, we first borrow the Rubin-Neyman potential outcome framework to define average causal effects of sensitive attributes. Then, we show that removing causal effects of sensitive attributes is equal to average counterfactual fairness in CF. Then, we use the propensity re-weighting paradigm to estimate the average causal effects of sensitive attributes and formulate the estimated causal effects as an additional regularization term. To the best of our knowledge, we are one of the first few attempts to achieve counterfactual fairness from the causal effect estimation perspective in CF, which frees us from building sophisticated causal graph. Finally, experiments on three real-world datasets show the superiority of our proposed model.

Document-level relation extraction (RE) aims to simultaneously predict relations (including no-relation cases denoted as NA) between all entity pairs in a document. It is typically formulated as a relation classification task with entities pre-detected in advance and solved by a hard-label training regime, which however neglects the divergence of the NA class and the correlations among other classes. This article introduces progressive self-distillation (PSD), a new training regime that employs online, self-knowledge distillation (KD) to produce and incorporate soft labels for document-level RE. The key idea of PSD is to gradually soften hard labels using past predictions from an RE model itself, which are adjusted adaptively as training proceeds. As such, PSD has to learn only one RE model within a single training pass, requiring no extra computation or annotation to pretrain another high-capacity teacher. PSD is conceptually simple, easy to implement, and generally applicable to various RE models to further improve their performance, without introducing additional parameters or significantly increasing training overheads into the models. It is also a general framework that can be flexibly extended to distilling various types of knowledge, rather than being restricted to soft labels themselves. Extensive experiments on four benchmarking datasets verify the effectiveness and generality of the proposed approach. The code is available at https://github.com/GaoJieCN/psd.