Sequential recommendation aims to predict the next item of interest for a user, based on her/his interaction history. In conventional sequential recommenders, a common approach is to learn sequence representations based on ID embeddings of items, which can be leveraged to predict the subsequent items of interest. Clearly, the sequence representations encode user behavioral patterns, which are critical to recommendation. Inspired by recent success in empowering Large Language Models (LLMs) to understand diverse modality (e.g., image, audio), a compelling question arises: “Can LLMs understand and utilize representations from conventional recommenders?” To answer this, we propose RecInterpreter, which examines the capacity of LLMs to decipher the representation space of pretrained recommenders. Specifically, with the multimodal pairs (i.e., interaction sequence representations and text narrations), RecInterpreter first uses a lightweight projector to map the representations into the token embedding space of the LLM, encouraging LLM to generate textual narrations for items within the sequence. Furthermore, upon interpreting recommenders, LLM can enhance its recommendation capabilities through fine-tuning with the projected representations, even without textual description of interaction sequences. Experiments showcase that RecInterpreter enhances LLMs to understand hidden representations from ID-based sequential recommenders and better accomplish recommendation task with the explicit understanding of behavior patterns.

The emergence of large generative models is transforming the landscape of recommender systems. One of the most fundamental components in building these models is action tokenization, the process of converting human-readable data (e.g., user-item interactions) into machine-readable formats (e.g., discrete token sequences). In this tutorial, we present a comprehensive overview of existing action tokenization techniques, converting actions to (1) item IDs, (2) textual descriptions, and (3) semantic IDs. We then make an in-depth discussion on the challenges and open questions of building large generative recommendation models from the perspective of action tokenization. Materials of this tutorial are available at: https://large-genrec.github.io/.

News recommender systems rely on rich textual information from news articles to generate user-specific recommendations. This reliance may expose these systems to potential vulnerabilities through textual attacks. To explore this vulnerability, we propose LANCE, a LArge language model-based News Content rEwriting framework, designed to influence news rankings and highlight the unintended promotion of manipulated news. LANCE consists of two key components: an explorer and a reflector. The explorer first generates rewritten news using diverse prompts, incorporating different writing styles, sentiments, and personas. We then collect these rewrites, evaluate their ranking impact within news recommender systems, and apply a filtering mechanism to retain effective rewrites. Next, the reflector fine-tunes an open-source LLM using the successful rewrites, enhancing its ability to generate more effective textual attacks. Experimental results demonstrate the effectiveness of LANCE in manipulating rankings within news recommender systems. Unlike attacks in other recomendation domains, negative and neutral rewrites consistently outperform positive ones, revealing a unique vulnerability specific to news recommendation. Once trained, LANCE successfully attacks unseen news recommender systems (i.e., those for which LANCE received no information during training), highlighting its generalization ability and exposing shared vulnerabilities across different systems. Our work underscores the urgent need for research on textual attacks and paves the way for future studies on defense strategies.

Diffusion models have shown significant potential in generating oracle items that best match user preference with guidance from user historical interaction sequences.However, the quality of guidance is often compromised by unpredictable missing data in observed sequence, leading to suboptimal item generation.Since missing data is uncertain in both occurrence and content, recovering it is impractical and may introduce additional errors.To tackle this challenge, we propose a novel dual-side Thompson sampling-based Diffusion Model (TDM), which simulates extra missing data in the guidance signals and allows diffusion models to handle existing missing data through extrapolation.To preserve user preference evolution in sequences despite extra missing data, we introduce Dual-side Thompson Sampling to implement simulation with two probability models, sampling by exploiting user preference from both item continuity and sequence stability.TDM strategically removes items from sequences based on dual-side Thompson sampling and treats these edited sequences as guidance for diffusion models, enhancing models' robustness to missing data through consistency regularization.Additionally, to enhance the generation efficiency, TDM is implemented under the denoising diffusion implicit models to accelerate the reverse process.Extensive experiments and theoretical analysis validate the effectiveness of TDM in addressing missing data in sequential recommendations.Our data and code is available at https

Collaborative recommendation fundamentally involves learning high-quality user and item representations from interaction data. Recently, graph convolution networks (GCNs) have advanced the field by utilizing high-order connectivity patterns in interaction graphs, as evidenced by state-of-the-art methods like PinSage and LightGCN. However, one key limitation has not been well addressed in existing solutions: capturing long-range collaborative filtering signals, which are crucial for modeling user preference. In this work, we propose a new graph transformer (GT) framework— Position-aware Graph Transformer for Recommendation (PGTR), which combines the global modeling capability of Transformer blocks with the local neighborhood feature extraction of GCNs. The key insight is to explicitly incorporate node position and structure information from the user-item interaction graph into GT architecture via several purpose-designed positional encodings. The long-range collaborative signals from the Transformer block are then combined linearly with the local neighborhood features from the GCN backbone to enhance node embeddings for final recommendations. Empirical studies demonstrate the effectiveness of the proposed PGTR method when implemented on various GCN-based backbones across four real-world datasets and the robustness against interaction sparsity as well as noise. Our implementations are available in GitHub: https://github.com/MEICRS/PGTR .

Designing effective prompts can empower LLMs to understand user preferences and provide recommendations with intent comprehension and knowledge utilization capabilities. Nevertheless, recent studies predominantly concentrate on task-wise prompting, developing fixed prompt templates shared across all users in a given recommendation task (e.g., rating or ranking). Although convenient, task-wise prompting overlooks individual user differences, leading to inaccurate analysis of user interests. In this work, we introduce the concept of instance-wise prompting, aiming at personalizing discrete prompts for individual users. Toward this end, we propose Reinforced Prompt Personalization (RPP) to realize it automatically. To improve efficiency and quality, RPP personalizes prompts at the sentence level rather than searching in the vast vocabulary word-by-word. Specifically, RPP breaks down the prompt into four patterns, tailoring patterns based on multi-agent and combining them. Then the personalized prompts interact with LLMs (environment) iteratively, to boost LLMs’ recommending performance (reward). In addition to RPP, to improve the scalability of action space, our proposal of RPP+ dynamically refines the selected actions with LLMs throughout the iterative process. Extensive experiments on various datasets demonstrate the superiority of RPP/RPP+ over traditional recommender models, few-shot methods, and other prompt-based methods, underscoring the significance of instance-wise prompting in LLMs for recommendation. Our code is available at https://github.com/maowenyu-11/RPP .

Recommendation unlearning is an emerging task to serve users for erasing unusable data (e.g., some historical behaviors) from a well-trained recommender model. Existing methods process unlearning requests by fully or partially retraining the model after removing the unusable data. However, these methods are impractical due to the high computation cost of full retraining and the highly possible performance damage of partial training. In this light, a desired recommendation unlearning method should obtain a similar model as full retraining in a more efficient manner, i.e., achieving complete, efficient and harmless unlearning. In this work, we propose a new Influence Function-based Recommendation Unlearning (IFRU) framework, which efficiently updates the model without retraining by estimating the influence of the unusable data on the model via the influence function . In the light that recent recommender models use historical data for both the constructions of the optimization loss and the computational graph (e.g., neighborhood aggregation), IFRU jointly estimates the direct influence of unusable data on optimization loss and the spillover influence on the computational graph to pursue complete unlearning. Furthermore, we propose an importance-based pruning algorithm to reduce the cost of the influence function. IFRU is harmless and applicable to mainstream differentiable models. Extensive experiments demonstrate that IFRU achieves more than 250 times acceleration compared to retraining-based methods with recommendation performance comparable to full retraining. Codes are available at https://github.com/baiyimeng/IFRU.

Masked modeling has recently achieved remarkable success in specific fields of vision and language, sparking a surge of interest in graph-related research. However, Masked Graph Modeling (MGM), which captures fine-grained local information by masking low-level elements such as nodes, edges, and features, limits itself to a sub-optimal position, particularly on tasks requiring high-quality graph-level representations. Such a local perspective disregards the graph's global information and structure. To address these limitations, we propose a novel graph pre-training framework called Graph Contrastive Masked Autoencoder (GCMAE). GCMAE leverages the strengths of both MGM and Graph Contrastive Learning (GCL) to provide a more comprehensive perspective of both local and global. Our frame-work uses instance discrimination to learn global representations of graphs and reconstructs the graph using masked low-level elements. We augment the framework with a novel multi-view augmentation module to further enhance the pre-trained model's robustness and generalization ability. We evaluate GCMAE on real-world biochemistry and social network datasets, conducting extensive experiments on both node and graph classification tasks and transfer learning on downstream graph classification tasks. Our experimental results demonstrate that GCMAE's comprehensive perspective of both local and global benefits model pre-training. Moreover, GCMAE outperforms existing MGM and GCL baselines, proving its effectiveness on downstream tasks. Our code is available at https://github.com/lyc0930/GCMAE.

Loss functions steer the optimization direction of recommendation models and are critical to model performance, but have received relatively little attention in recent recommendation research. Among various losses, we find Softmax loss (SL) stands out for not only achieving remarkable accuracy but also better robustness and fairness. Nevertheless, the current literature lacks a comprehensive explanation for the efficacy of SL. Toward addressing this research gap, we conduct theoretical analyses on SL and uncover three insights: 1) Optimizing SL is equivalent to performing Distributionally Robust Optimization (DRO) on the negative data, thereby learning against perturbations on the negative distribution and yielding robustness to noisy negatives. 2) Comparing with other loss functions, SL implicitly penalizes the prediction variance, resulting in a smaller gap between predicted values and and thus producing fairer results. Building on these insights, we further propose a novel loss function Bilateral SoftMax Loss (BSL) that extends the advantage of SL to both positive and negative sides. BSL augments SL by applying the same Log-Expectation-Exp structure to positive examples as is used for negatives, making the model robust to the noisy positives as well. Remarkably, BSL is simple and easy-to-implement - requiring just one additional line of code compared to SL. Experiments on four real-world datasets and three representative backbones demonstrate the effectiveness of our proposal. The code is available at https://github.com/junkangwu/BSL.

Sequential recommendation aims to predict users' next interaction with items based on their past engagement sequence. Recently, the advent of Large Language Models (LLMs) has sparked interest in leveraging them for sequential recommendation, viewing it as language modeling. Previous studies represent items within LLMs' input prompts as either ID indices or textual metadata. However, these approaches often fail to either encapsulate comprehensive world knowledge or exhibit sufficient behavioral understanding. To combine the complementary strengths of conventional recommenders in capturing behavioral patterns of users and LLMs in encoding world knowledge about items, we introduce Large Language-Recommendation Assistant (LLaRA). Specifically, it uses a novel hybrid prompting method that integrates ID-based item embeddings learned by traditional recommendation models with textual item features. Treating the "sequential behaviors of users" as a distinct modality beyond texts, we employ a projector to align the traditional recommender's ID embeddings with the LLM's input space. Moreover, rather than directly exposing the hybrid prompt to LLMs, a curriculum learning strategy is adopted to gradually ramp up training complexity. Initially, we warm up the LLM using text-only prompts, which better suit its inherent language modeling ability. Subsequently, we progressively transition to the hybrid prompts, training the model to seamlessly incorporate the behavioral knowledge from the traditional sequential recommender into the LLM. Empirical results validate the effectiveness of our proposed framework. Codes are available at https://github.com/ljy0ustc/LLaRA.

Medication recommendation systems have gained significant attention in healthcare as a means of providing tailored and effective drug combinations based on patients' clinical information. However, existing approaches often suffer from fairness issues, as recommendations tend to be more accurate for patients with common diseases compared to those with rare conditions. In this paper, we propose a novel model called Robust and Accurate REcommendations for Medication (RAREMed), which leverages the pretrain-finetune learning paradigm to enhance accuracy for rare diseases. RAREMed employs a transformer encoder with a unified input sequence approach to capture complex relationships among disease and procedure codes. Additionally, it introduces two self-supervised pre-training tasks, namely Sequence Matching Prediction (SMP) and Self Reconstruction (SR), to learn specialized medication needs and interrelations among clinical codes. Experimental results on two real-world datasets demonstrate that RAREMed provides accurate drug sets for both rare and common disease patients, thereby mitigating unfairness in medication recommendation systems. The implementation is available via https://github.com/zzhUSTC2016/RAREMed.

Conventional recommender systems (RSs) face challenges in precisely capturing users' fine-grained preferences. Large language models (LLMs) have shown capabilities in commonsense reasoning and leveraging external tools that may help address these challenges. However, existing LLM-based RSs suffer from hallucinations, misalignment between the semantic space of items and the behavior space of users, or overly simplistic control strategies (e.g., whether to rank or directly present existing results). To bridge these gap, we introduce ToolRec, a framework for LLM-empowered recommendations via tool learning that uses LLMs as surrogate users, thereby guiding the recommendation process and invoking external tools to generate a recommendation list that aligns closely with users' nuanced preferences.

In the realm of deep learning-based recommendation systems, the increasing computational demands, driven by the growing number of users and items, pose a significant challenge to practical deployment. This challenge is primarily twofold: reducing the model size while effectively learning user and item representations for efficient recommendations. Despite considerable advancements in model compression and architecture search, prevalent approaches face notable constraints. These include substantial additional computational costs from pre-training/re-training in model compression and an extensive search space in architecture design. Additionally, managing complexity and adhering to memory constraints is problematic, especially in scenarios with strict time or space limitations. Addressing these issues, this paper introduces a novel learning paradigm, Dynamic Sparse Learning (DSL), tailored for recommendation models. DSL innovatively trains a lightweight sparse model from scratch, periodically evaluating and dynamically adjusting each weight's significance and the model's sparsity distribution during the training. This approach ensures a consistent and minimal parameter budget throughout the full learning lifecycle, paving the way for "end-to-end" efficiency from training to inference. Our extensive experimental results underline DSL's effectiveness, significantly reducing training and inference costs while delivering comparable recommendation performance. We give an code link of our work: https://github.com/shuyao-wang/DSL.

In graph classification, attention- and pooling-based graph neural networks (GNNs) predominate to extract salient features from the input graph and support the prediction. They mostly follow the paradigm of “learning to attend,” which maximizes the mutual information between the attended graph and the ground-truth label. However, this paradigm causes GNN classifiers to indiscriminately absorb all statistical correlations between input features and labels in the training data without distinguishing the causal and noncausal effects of features. Rather than emphasizing causal features, the attended graphs tend to rely on noncausal features as shortcuts to predictions. These shortcut features may easily change outside the training distribution, thereby leading to poor generalization for GNN classifiers. In this article, we take a causal view on GNN modeling. Under our causal assumption, the shortcut feature serves as a confounder between the causal feature and prediction. It misleads the classifier into learning spurious correlations that facilitate prediction in in-distribution (ID) test evaluation while causing significant performance drop in out-of-distribution (OOD) test data. To address this issue, we employ the backdoor adjustment from causal theory—combining each causal feature with various shortcut features, to identify causal patterns and mitigate the confounding effect. Specifically, we employ attention modules to estimate the causal and shortcut features of the input graph. Then, a memory bank collects the estimated shortcut features, enhancing the diversity of shortcut features for combination. Simultaneously, we apply the prototype strategy to improve the consistency of intra-class causal features. We term our method as CAL+, which can promote stable relationships between causal estimation and prediction, regardless of distribution changes. Extensive experiments on synthetic and real-world OOD benchmarks demonstrate our method’s effectiveness in improving OOD generalization. Our codes are released at https://github.com/shuyao-wang/CAL-plus .

The learning objective plays a fundamental role to build a recommender system. Most methods routinely adopt either pointwise (e.g., binary cross-entropy) or pairwise (e.g., BPR) loss to train the model parameters, while rarely pay attention to softmax loss, which assumes the probabilities of all classes sum up to 1, due to its computational complexity when scaling up to large datasets or intractability for streaming data where the complete item space is not always available. The sampled softmax (SSM) loss emerges as an efficient substitute for softmax loss. Its special case, InfoNCE loss, has been widely used in self-supervised learning and exhibited remarkable performance for contrastive learning. Nonetheless, limited recommendation work uses the SSM loss as the learning objective. Worse still, none of them explores its properties thoroughly and answers “Does SSM loss suit for item recommendation?” and “What are the conceptual advantages of SSM loss, as compared with the prevalent losses?”, to the best of our knowledge. In this work, we aim at offering a better understanding of SSM for item recommendation. Specifically, we first theoretically reveal three model-agnostic advantages: (1) mitigating popularity bias, which is beneficial to long-tail recommendation; (2) mining hard negative samples, which offers informative gradients to optimize model parameters; and (3) maximizing the ranking metric, which facilitates top- K performance. However, based on our empirical studies, we recognize that the default choice of cosine similarity function in SSM limits its ability in learning the magnitudes of representation vectors. As such, the combinations of SSM with the models that also fall short in adjusting magnitudes (e.g., matrix factorization) may result in poor representations. One step further, we provide mathematical proof that message passing schemes in graph convolution networks can adjust representation magnitude according to node degree, which naturally compensates for the shortcoming of SSM. Extensive experiments on four benchmark datasets justify our analyses, demonstrating the superiority of SSM for item recommendation. Our implementations are available in both TensorFlow 1 and PyTorch. 2

Leading sequential recommendation (SeqRec) models adopt empirical risk minimization (ERM) as the learning framework, which inherently assumes that the training data (historical interaction sequences) and the testing data (future interactions) are drawn from the same distribution. However, such i.i.d. assumption hardly holds in practice, due to the online serving and dynamic nature of recommender system.For example, with the streaming of new data, the item popularity distribution would change, and the user preference would evolve after consuming some items. Such distribution shifts could undermine the ERM framework, hurting the model's generalization ability for future online serving.

Recent years have witnessed the great successes of embedding-based methods in recommender systems. Despite their decent performance, we argue one potential limitation of these methods — the embedding magnitude has not been explicitly modulated, which may aggravate popularity bias and training instability, hindering the model from making a good recommendation. It motivates us to leverage the embedding normalization in recommendation. By normalizing user/item embeddings to a specific value, we empirically observe impressive performance gains (9% on average) on four real-world datasets. Although encouraging, we also reveal a serious limitation when applying normalization in recommendation — the performance is highly sensitive to the choice of the temperature τ which controls the scale of the normalized embeddings.

With the greater emphasis on privacy and security in our society, the problem of graph unlearning — revoking the influence of specific data on the trained GNN model, is drawing increasing attention. However, ranging from machine unlearning to recently emerged graph unlearning methods, existing efforts either resort to retraining paradigm, or perform approximate erasure that fails to consider the inter-dependency between connected neighbors or imposes constraints on GNN structure, therefore hard to achieve satisfying performance-complexity trade-offs.

In graph classification, attention- and pooling-based graph neural networks (GNNs) prevail to extract the critical features from the input graph and support the prediction. They mostly follow the paradigm of learning to attend, which maximizes the mutual information between the attended graph and the ground-truth label. However, this paradigm makes GNN classifiers recklessly absorb all the statistical correlations between input features and labels in the training data, without distinguishing the causal and noncausal effects of features. Instead of underscoring the causal features, the attended graphs are prone to visit the noncausal features as the shortcut to predictions. Such shortcut features might easily change outside the training distribution, thereby making the GNN classifiers suffer from poor generalization.

Most recommender systems optimize the model on observed interaction data, which is affected by the previous exposure mechanism and exhibits many biases like popularity bias. The loss functions, such as the mostly used pointwise Binary Cross-Entropy and pairwise Bayesian Personalized Ranking, are not designed to consider the biases in observed data. As a result, the model optimized on the loss would inherit the data biases, or even worse, amplify the biases. For example, a few popular items take up more and more exposure opportunities, severely hurting the recommendation quality on niche items — known as the notorious Mathew effect.

Representation learning on user-item graph for recommendation has evolved from using single ID or interaction history to exploiting higher-order neighbors. This leads to the success of graph convolution networks (GCNs) for recommendation such as PinSage and LightGCN. Despite effectiveness, we argue that they suffer from two limitations: (1) high-degree nodes exert larger impact on the representation learning, deteriorating the recommendations of low-degree (long-tail) items; and (2) representations are vulnerable to noisy interactions, as the neighborhood aggregation scheme further enlarges the impact of observed edges.