本文精选了上周（0501-0507）最新发布的16篇推荐系统相关论文，主要研究方向包括推荐系统中的偏差问题、强化学习对话推荐、公平性问题、大型语言模型赋能推荐系统、鲁棒推荐系统、多模态推荐等。
以下整理了论文标题以及摘要，如感兴趣可移步原文精读。
1. Maximizing Submodular Functions for Recommendation in the Presence of Biases, WWW2023
2. Towards Hierarchical Policy Learning for Conversational Recommendation with Hypergraph-based Reinforcement Learning, IJCAI2023
3. Inference at Scale: Significance Testing for Large Search and Recommendation Experiments, SIGIR2023
4. When Newer is Not Better: Does Deep Learning Really Benefit Recommendation From Implicit Feedback? SIGIR2023
5. Multidimensional Fairness in Paper Recommendation
6. Leveraging Language Representation for Material Recommendation, Ranking, and Exploration
7. Explicit Knowledge Graph Reasoning for Conversational Recommendation
8. TALLRec: An Effective and Efficient Tuning Framework to Align Large Language Model with Recommendation
9. Beyond Prediction: On-street Parking Recommendation using Heterogeneous Graph-based List-wise Ranking
10. Knowledge-refined Denoising Network for Robust Recommendation, SIGIR2023
11. Ensemble Modeling with Contrastive Knowledge Distillation for Sequential Recommendation, SIGIR2023
12. Disentangled Contrastive Collaborative Filtering, SIGIR2023
13. Uncovering ChatGPT's Capabilities in Recommender Systems
14. Denoising Multi-modal Sequential Recommenders with Contrastive Learning
15. Structure Aware Incremental Learning with Personalized Imitation Weights for Recommender Systems
16. The Dark Side of Explanations: Poisoning Recommender Systems with Counterfactual Examples, SIGIR2023
1. Maximizing Submodular Functions for Recommendation in the Presence of Biases, WWW2023
Anay Mehrotra, Nisheeth K. Vishnoi
https://arxiv.org/abs/2305.02806
Subset selection tasks, arise in recommendation systems and search engines and ask to select a subset of items that maximize the value for the user. The values of subsets often display diminishing returns, and hence, submodular functions have been used to model them. If the inputs defining the submodular function are known, then existing algorithms can be used. In many applications, however, inputs have been observed to have social biases that reduce the utility of the output subset. Hence, interventions to improve the utility are desired. Prior works focus on maximizing linear functions -- a special case of submodular functions -- and show that fairness constraint-based interventions can not only ensure proportional representation but also achieve near-optimal utility in the presence of biases. We study the maximization of a family of submodular functions that capture functions arising in the aforementioned applications. Our first result is that, unlike linear functions, constraint-based interventions cannot guarantee any constant fraction of the optimal utility for this family of submodular functions. Our second result is an algorithm for submodular maximization. The algorithm provably outputs subsets that have near-optimal utility for this family under mild assumptions and that proportionally represent items from each group. In empirical evaluation, with both synthetic and real-world data, we observe that this algorithm improves the utility of the output subset for this family of submodular functions over baselines.
2. Towards Hierarchical Policy Learning for Conversational Recommendation with Hypergraph-based Reinforcement Learning, IJCAI2023
Sen Zhao1, Wei Wei, Yifan Liu, Ziyang Wang, Wendi Li, Xian-Ling Mao, Shuai Zhu, Minghui Yang, Zujie Wen
https://arxiv.org/abs/2305.02575
Conversational recommendation systems (CRS) aim to timely and proactively acquire user dynamic preferred attributes through conversations for item recommendation. In each turn of CRS, there naturally have two decision-making processes with different roles that influence each other: 1) director, which is to select the follow-up option (i.e., ask or recommend) that is more effective for reducing the action space and acquiring user preferences; and 2) actor, which is to accordingly choose primitive actions (i.e., asked attribute or recommended item) that satisfy user preferences and give feedback to estimate the effectiveness of the director's option. However, existing methods heavily rely on a unified decision-making module or heuristic rules, while neglecting to distinguish the roles of different decision procedures, as well as the mutual influences between them. To address this, we propose a novel Director-Actor Hierarchical Conversational Recommender (DAHCR), where the director selects the most effective option, followed by the actor accordingly choosing primitive actions that satisfy user preferences. Specifically, we develop a dynamic hypergraph to model user preferences and introduce an intrinsic motivation to train from weak supervision over the director. Finally, to alleviate the bad effect of model bias on the mutual influence between the director and actor, we model the director's option by sampling from a categorical distribution. Extensive experiments demonstrate that DAHCR outperforms state-of-the-art methods.
3. Inference at Scale: Significance Testing for Large Search and Recommendation Experiments, SIGIR2023
Ngozi Ihemelandu, Michael D. Ekstrand
https://arxiv.org/abs/2305.02461
A number of information retrieval studies have been done to assess which statistical techniques are appropriate for comparing systems. However, these studies are focused on TREC-style experiments, which typically have fewer than 100 topics. There is no similar line of work for large search and recommendation experiments; such studies typically have thousands of topics or users and much sparser relevance judgements, so it is not clear if recommendations for analyzing traditional TREC experiments apply to these settings. In this paper, we empirically study the behavior of significance tests with large search and recommendation evaluation data. Our results show that the Wilcoxon and Sign tests show significantly higher Type-1 error rates for large sample sizes than the bootstrap, randomization and t-tests, which were more consistent with the expected error rate. While the statistical tests displayed differences in their power for smaller sample sizes, they showed no difference in their power for large sample sizes. We recommend the sign and Wilcoxon tests should not be used to analyze large scale evaluation results. Our result demonstrate that with Top-N recommendation and large search evaluation data, most tests would have a 100% chance of finding statistically significant results. Therefore, the effect size should be used to determine practical or scientific significance.
4. When Newer is Not Better: Does Deep Learning Really Benefit Recommendation From Implicit Feedback? SIGIR2023
Yushun Dong, Jundong Li, Tobias Schnabel
https://arxiv.org/abs/2305.01801
In recent years, neural models have been repeatedly touted to exhibit state-of-the-art performance in recommendation. Nevertheless, multiple recent studies have revealed that the reported state-of-the-art results of many neural recommendation models cannot be reliably replicated. A primary reason is that existing evaluations are performed under various inconsistent protocols. Correspondingly, these replicability issues make it difficult to understand how much benefit we can actually gain from these neural models. It then becomes clear that a fair and comprehensive performance comparison between traditional and neural models is needed.
Motivated by these issues, we perform a large-scale, systematic study to compare recent neural recommendation models against traditional ones in top-n recommendation from implicit data. We propose a set of evaluation strategies for measuring memorization performance, generalization performance, and subgroup-specific performance of recommendation models. We conduct extensive experiments with 13 popular recommendation models (including two neural models and 11 traditional ones as baselines) on nine commonly used datasets. Our experiments demonstrate that even with extensive hyper-parameter searches, neural models do not dominate traditional models in all aspects, e.g., they fare worse in terms of average HitRate. We further find that there are areas where neural models seem to outperform non-neural models, for example, in recommendation diversity and robustness between different subgroups of users and items. Our work illuminates the relative advantages and disadvantages of neural models in recommendation and is therefore an important step towards building better recommender systems.
5. Multidimensional Fairness in Paper Recommendation
Reem Alsaffar, Susan Gauch, Hiba Al-Kawaz
https://arxiv.org/abs/2305.01141
To prevent potential bias in the paper review and selection process for conferences and journals, most include double blind review. Despite this, studies show that bias still exists. Recommendation algorithms for paper review also may have implicit bias. We offer three fair methods that specifically take into account author diversity in paper recommendation to address this. Our methods provide fair outcomes across many protected variables concurrently, in contrast to typical fair algorithms that only use one protected variable. Five demographic characteristics-gender, ethnicity, career stage, university rank, and geolocation-are included in our multidimensional author profiles. The Overall Diversity approach uses a score for overall diversity to rank publications. The Round Robin Diversity technique chooses papers from authors who are members of each protected group in turn, whereas the Multifaceted Diversity method chooses papers that initially fill the demographic feature with the highest importance. We compare the effectiveness of author diversity profiles based on Boolean and continuous-valued features. By selecting papers from a pool of SIGCHI 2017, DIS 2017, and IUI 2017 papers, we recommend papers for SIGCHI 2017 and evaluate these algorithms using the user profiles. We contrast the papers that were recommended with those that were selected by the conference. We find that utilizing profiles with either Boolean or continuous feature values, all three techniques boost diversity while just slightly decreasing utility or not decreasing. By choosing authors who are 42.50% more diverse and with a 2.45% boost in utility, our best technique, Multifaceted Diversity, suggests a set of papers that match demographic parity. The selection of grant proposals, conference papers, journal articles, and other academic duties might all use this strategy.
6. Leveraging Language Representation for Material Recommendation, Ranking, and Exploration
Jiaxing Qu, Yuxuan Richard Xie, Elif Ertekin
https://arxiv.org/abs/2305.01101
Data-driven approaches for material discovery and design have been accelerated by emerging efforts in machine learning. While there is enormous progress towards learning the structure to property relationship of materials, methods that allow for general representations of crystals to effectively explore the vast material search space and identify high-performance candidates remain limited. In this work, we introduce a material discovery framework that uses natural language embeddings derived from material science-specific language models as representations of compositional and structural features. The discovery framework consists of a joint scheme that, given a query material, first recalls candidates based on representational similarity, and ranks the candidates based on target properties through multi-task learning. The contextual knowledge encoded in language representations is found to convey information about material properties and structures, enabling both similarity analysis for recall, and multi-task learning to share information for related properties. By applying the discovery framework to thermoelectric materials, we demonstrate diversified recommendations of prototype structures and identify under-studied high-performance material spaces, including halide perovskite, delafossite-like, and spinel-like structures. By leveraging material language representations, our framework provides a generalized means for effective material recommendation, which is task-agnostic and can be applied to various material systems.
7. Explicit Knowledge Graph Reasoning for Conversational Recommendation
Xuhui Ren, Tong Chen, Quoc Viet Hung Nguyen, Lizhen Cui, Zi Huang, Hongzhi Yin
https://arxiv.org/abs/2305.00783
Traditional recommender systems estimate user preference on items purely based on historical interaction records, thus failing to capture fine-grained yet dynamic user interests and letting users receive recommendation only passively. Recent conversational recommender systems (CRSs) tackle those limitations by enabling recommender systems to interact with the user to obtain her/his current preference through a sequence of clarifying questions. Despite the progress achieved in CRSs, existing solutions are far from satisfaction in the following two aspects: 1) current CRSs usually require each user to answer a quantity of clarifying questions before reaching the final recommendation, which harms the user experience; 2) there is a semantic gap between the learned representations of explicitly mentioned attributes and items. To address these drawbacks, we introduce the knowledge graph (KG) as the auxiliary information for comprehending and reasoning a user's preference, and propose a new CRS framework, namely Knowledge Enhanced Conversational Reasoning (KECR) system. As a user can reflect her/his preference via both attribute- and item-level expressions, KECR closes the semantic gap between two levels by embedding the structured knowledge in the KG. Meanwhile, KECR utilizes the connectivity within the KG to conduct explicit reasoning of the user demand, making the model less dependent on the user's feedback to clarifying questions. KECR can find a prominent reasoning chain to make the recommendation explainable and more rationale, as well as smoothen the conversation process, leading to better user experience and conversational recommendation accuracy. Extensive experiments on two real-world datasets demonstrate our approach's superiority over state-of-the-art baselines in both automatic evaluations and human judgments.
8. TALLRec: An Effective and Efficient Tuning Framework to Align Large Language Model with Recommendation
Keqin Bao, Jizhi Zhang, Yang Zhang, Wenjie Wang, Fuli Feng, Xiangnan He
https://arxiv.org/abs/2305.00447
Large Language Models (LLMs) have demonstrated remarkable performance across diverse domains, thereby prompting researchers to explore their potential for use in recommendation systems. Initial attempts have leveraged the exceptional capabilities of LLMs, such as rich knowledge and strong generalization through In-context Learning, which involves phrasing the recommendation task as prompts. Nevertheless, the performance of LLMs in recommendation tasks remains suboptimal due to a substantial disparity between the training tasks for LLMs and recommendation tasks, as well as inadequate recommendation data during pre-training. To bridge the gap, we consider building a Large Recommendation Language Model by tunning LLMs with recommendation data. To this end, we propose an efficient and effective Tuning framework for Aligning LLMs with Recommendation, namely TALLRec. We have demonstrated that the proposed TALLRec framework can significantly enhance the recommendation capabilities of LLMs in the movie and book domains, even with a limited dataset of fewer than 100 samples. Additionally, the proposed framework is highly efficient and can be executed on a single RTX 3090 with LLaMA-7B. Furthermore, the fine-tuned LLM exhibits robust cross-domain generalization. Our code and data are available at: https://github.com/SAI990323/TALLRec
9. Beyond Prediction: On-street Parking Recommendation using Heterogeneous Graph-based List-wise Ranking
Hanyu Sun, Xiao Huang, Wei Ma
https://arxiv.org/abs/2305.00162
To provide real-time parking information, existing studies focus on predicting parking availability, which seems an indirect approach to saving drivers' cruising time. In this paper, we first time propose an on-street parking recommendation (OPR) task to directly recommend a parking space for a driver. To this end, a learn-to-rank (LTR) based OPR model called OPR-LTR is built. Specifically, parking recommendation is closely related to the "turnover events" (state switching between occupied and vacant) of each parking space, and hence we design a highly efficient heterogeneous graph called ESGraph to represent historical and real-time meters' turnover events as well as geographical relations; afterward, a convolution-based event-then-graph network is used to aggregate and update representations of the heterogeneous graph. A ranking model is further utilized to learn a score function that helps recommend a list of ranked parking spots for a specific on-street parking query. The method is verified using the on-street parking meter data in Hong Kong and San Francisco. By comparing with the other two types of methods: prediction-only and prediction-then-recommendation, the proposed direct-recommendation method achieves satisfactory performance in different metrics. Extensive experiments also demonstrate that the proposed ESGraph and the recommendation model are more efficient in terms of computational efficiency as well as saving drivers' on-street parking time.
10. Knowledge-refined Denoising Network for Robust Recommendation, SIGIR2023
Xinjun Zhu, Yuntao Du, Yuren Mao, Lu Chen, Yujia Hu, Yunjun Gao
https://arxiv.org/abs/2304.14987
Knowledge graph (KG), which contains rich side information, becomes an essential part to boost the recommendation performance and improve its explainability. However, existing knowledge-aware recommendation methods directly perform information propagation on KG and user-item bipartite graph, ignoring the impacts of task-irrelevant knowledge propagation and vulnerability to interaction noise, which limits their performance. To solve these issues, we propose a robust knowledge-aware recommendation framework, called Knowledge-refined Denoising Network (KRDN), to prune the task-irrelevant knowledge associations and noisy implicit feedback simultaneously. KRDN consists of an adaptive knowledge refining strategy and a contrastive denoising mechanism, which are able to automatically distill high-quality KG triplets for aggregation and prune noisy implicit feedback respectively. Besides, we also design the self-adapted loss function and the gradient estimator for model optimization. The experimental results on three benchmark datasets demonstrate the effectiveness and robustness of KRDN over the state-of-the-art knowledge-aware methods like KGIN, MCCLK, and KGCL, and also outperform robust recommendation models like SGL and SimGCL.
11. Ensemble Modeling with Contrastive Knowledge Distillation for Sequential Recommendation, SIGIR2023
Hanwen Du, Huanhuan Yuan, Pengpeng Zhao, Fuzhen Zhuang, Guanfeng Liu, Lei Zhao, Yanchi Liu, Victor S. Sheng
https://arxiv.org/abs/2304.14668
Sequential recommendation aims to capture users' dynamic interest and predicts the next item of users' preference. Most sequential recommendation methods use a deep neural network as sequence encoder to generate user and item representations. Existing works mainly center upon designing a stronger sequence encoder. However, few attempts have been made with training an ensemble of networks as sequence encoders, which is more powerful than a single network because an ensemble of parallel networks can yield diverse prediction results and hence better accuracy. In this paper, we present Ensemble Modeling with contrastive Knowledge Distillation for sequential recommendation (EMKD). Our framework adopts multiple parallel networks as an ensemble of sequence encoders and recommends items based on the output distributions of all these networks. To facilitate knowledge transfer between parallel networks, we propose a novel contrastive knowledge distillation approach, which performs knowledge transfer from the representation level via Intra-network Contrastive Learning (ICL) and Cross-network Contrastive Learning (CCL), as well as Knowledge Distillation (KD) from the logits level via minimizing the Kullback-Leibler divergence between the output distributions of the teacher network and the student network. To leverage contextual information, we train the primary masked item prediction task alongside the auxiliary attribute prediction task as a multi-task learning scheme. Extensive experiments on public benchmark datasets show that EMKD achieves a significant improvement compared with the state-of-the-art methods. Besides, we demonstrate that our ensemble method is a generalized approach that can also improve the performance of other sequential recommenders. Our code is available at this link: https://github.com/hw-du/EMKD
12. Disentangled Contrastive Collaborative Filtering, SIGIR2023
Xubin Ren, Lianghao Xia, Jiashu Zhao, Dawei Yin, Chao Huang
https://arxiv.org/abs/2305.02759
Recent studies show that graph neural networks (GNNs) are prevalent to model high-order relationships for collaborative filtering (CF). Towards this research line, graph contrastive learning (GCL) has exhibited powerful performance in addressing the supervision label shortage issue by learning augmented user and item representations. While many of them show their effectiveness, two key questions still remain unexplored: i) Most existing GCL-based CF models are still limited by ignoring the fact that user-item interaction behaviors are often driven by diverse latent intent factors (e.g., shopping for family party, preferred color or brand of products); ii) Their introduced non-adaptive augmentation techniques are vulnerable to noisy information, which raises concerns about the model's robustness and the risk of incorporating misleading self-supervised signals. In light of these limitations, we propose a Disentangled Contrastive Collaborative Filtering framework (DCCF) to realize intent disentanglement with self-supervised augmentation in an adaptive fashion. With the learned disentangled representations with global context, our DCCF is able to not only distill finer-grained latent factors from the entangled self-supervision signals but also alleviate the augmentation-induced noise. Finally, the cross-view contrastive learning task is introduced to enable adaptive augmentation with our parameterized interaction mask generator. Experiments on various public datasets demonstrate the superiority of our method compared to existing solutions. Our model implementation is released at the link https://github.com/HKUDS/DCCF
13. Uncovering ChatGPT's Capabilities in Recommender Systems
Sunhao Dai, Ninglu Shao, Haiyuan Zhao, Weijie Yu, Zihua Si, Chen Xu, Zhongxiang Sun, Xiao Zhang, Jun Xu
https://arxiv.org/abs/2305.02182
The debut of ChatGPT has recently attracted the attention of the natural language processing (NLP) community and beyond. Existing studies have demonstrated that ChatGPT shows significant improvement in a range of downstream NLP tasks, but the capabilities and limitations of ChatGPT in terms of recommendations remain unclear. In this study, we aim to conduct an empirical analysis of ChatGPT's recommendation ability from an Information Retrieval (IR) perspective, including point-wise, pair-wise, and list-wise ranking. To achieve this goal, we re-formulate the above three recommendation policies into a domain-specific prompt format. Through extensive experiments on four datasets from different domains, we demonstrate that ChatGPT outperforms other large language models across all three ranking policies. Based on the analysis of unit cost improvements, we identify that ChatGPT with list-wise ranking achieves the best trade-off between cost and performance compared to point-wise and pair-wise ranking. Moreover, ChatGPT shows the potential for mitigating the cold start problem and interpretable recommendation. To facilitate further explorations in this area, the full code and detailed original results are open-sourced at https://github.com/rainym00d/LLM4RS
14. Denoising Multi-modal Sequential Recommenders with Contrastive Learning
Dong Yao, Shengyu Zhang, Zhou Zhao, Jieming Zhu, Wenqiao Zhang, Rui Zhang, Xiaofei He, Fei Wu
https://arxiv.org/abs/2305.01915
There is a rapidly-growing research interest in engaging users with multi-modal data for accurate user modeling on recommender systems. Existing multimedia recommenders have achieved substantial improvements by incorporating various modalities and devising delicate modules. However, when users decide to interact with items, most of them do not fully read the content of all modalities. We refer to modalities that directly cause users' behaviors as point-of-interests, which are important aspects to capture users' interests. In contrast, modalities that do not cause users' behaviors are potential noises and might mislead the learning of a recommendation model. Not surprisingly, little research in the literature has been devoted to denoising such potential noises due to the inaccessibility of users' explicit feedback on their point-of-interests. To bridge the gap, we propose a weakly-supervised framework based on contrastive learning for denoising multi-modal recommenders (dubbed Demure). In a weakly-supervised manner, Demure circumvents the requirement of users' explicit feedback and identifies the noises by analyzing the modalities of all interacted items from a given user.
15. Structure Aware Incremental Learning with Personalized Imitation Weights for Recommender Systems
Yuening Wang, Yingxue Zhang, Antonios Valkanas, Ruiming Tang, Chen Ma, Jianye Hao, Mark Coates
https://arxiv.org/abs/2305.01204
Recommender systems now consume large-scale data and play a significant role in improving user experience. Graph Neural Networks (GNNs) have emerged as one of the most effective recommender system models because they model the rich relational information. The ever-growing volume of data can make training GNNs prohibitively expensive. To address this, previous attempts propose to train the GNN models incrementally as new data blocks arrive. Feature and structure knowledge distillation techniques have been explored to allow the GNN model to train in a fast incremental fashion while alleviating the catastrophic forgetting problem. However, preserving the same amount of the historical information for all users is sub-optimal since it fails to take into account the dynamics of each user's change of preferences. For the users whose interests shift substantially, retaining too much of the old knowledge can overly constrain the model, preventing it from quickly adapting to the users' novel interests. In contrast, for users who have static preferences, model performance can benefit greatly from preserving as much of the user's long-term preferences as possible. In this work, we propose a novel training strategy that adaptively learns personalized imitation weights for each user to balance the contribution from the recent data and the amount of knowledge to be distilled from previous time periods. We demonstrate the effectiveness of learning imitation weights via a comparison on five diverse datasets for three state-of-art structure distillation based recommender systems. The performance shows consistent improvement over competitive incremental learning techninques.
16. The Dark Side of Explanations: Poisoning Recommender Systems with Counterfactual Examples, SIGIR2023
Ziheng Chen, Fabrizio Silvestri, Jia Wang, Yongfeng Zhang, Gabriele Tolomei
https://arxiv.org/abs/2305.00574
Deep learning-based recommender systems have become an integral part of several online platforms. However, their black-box nature emphasizes the need for explainable artificial intelligence (XAI) approaches to provide human-understandable reasons why a specific item gets recommended to a given user. One such method is counterfactual explanation (CF). While CFs can be highly beneficial for users and system designers, malicious actors may also exploit these explanations to undermine the system's security. In this work, we propose H-CARS, a novel strategy to poison recommender systems via CFs. Specifically, we first train a logical-reasoning-based surrogate model on training data derived from counterfactual explanations. By reversing the learning process of the recommendation model, we thus develop a proficient greedy algorithm to generate fabricated user profiles and their associated interaction records for the aforementioned surrogate model. Our experiments, which employ a well-known CF generation method and are conducted on two distinct datasets, show that H-CARS yields significant and successful attack performance.
