Search

The article focuses on improving generative retrieval (GR) models by introducing a listwise approach that optimizes relevance at the document list level, unlike existing models that use pointwise approaches. It mentions by treating the generation of a ranked document list as a sequence learning process, the proposed method maximizes the generation likelihood of each document given the preceding documents in the list, addressing the sub-optimality of pointwise approaches.

The article introduces a dual meta-learning (DML) technique to enhance semi-supervised text classification, improving both teacher and student classifiers iteratively. Topics include the challenge of noisy pseudo-labels in semi-supervised text classification, the proposed meta-learning methods for noise correction and pseudo supervision, and the experimental validation demonstrating the effectiveness of the DML framework.

Domain generalization aims to design models that can effectively generalize to unseen target domains by learning from observed source domains. Domain generalization poses a significant challenge for time series data, due to varying data distributions and temporal dependencies. Existing approaches to domain generalization are not designed for time series data, which often results in suboptimal or unstable performance when confronted with diverse temporal patterns and complex data characteristics. We propose a novel approach to tackle the problem of domain generalization in time series forecasting. We focus on a scenario where time series domains share certain common attributes and exhibit no abrupt distribution shifts. Our method revolves around the incorporation of a key regularization term into an existing time series forecasting model: domain discrepancy regularization. In this way, we aim to enforce consistent performance across different domains that exhibit distinct patterns. We calibrate the regularization term by investigating the performance within individual domains and propose the domain discrepancy regularization with domain difficulty awareness. We demonstrate the effectiveness of our method on multiple datasets, including synthetic and real-world time series datasets from diverse domains such as retail, transportation, and finance. Our method is compared against traditional methods, deep learning models, and domain generalization approaches to provide comprehensive insights into its performance. In these experiments, our method showcases superior performance, surpassing both the base model and competing domain generalization models across all datasets. Furthermore, our method is highly general and can be applied to various time series models. [ABSTRACT FROM AUTHOR]

The article investigates user satisfaction in conversational recommender systems (CRSs) and proposes a method for understanding and predicting it. It explores the impact of various dialogue aspects on user satisfaction, including relevance, interestingness, understanding, task completion, interest arousal, and efficiency. It is reported that unlike previous methods relying on turn-level satisfaction ratings, the proposed approach considers broader dialogue aspects.

The article focuses on improving the evaluation of task-oriented dialogue systems (TDSs) through the use of user simulators. Topics include the limitations of current TDS evaluation methods, the challenges of employing user simulators for TDS evaluation, and the proposed solution which involves a metaphorical user simulator (MetaSim) and a tester-based evaluation framework.

The article focuses on evaluating Next Basket Recommendation (NBR) methods for suggesting items in users future shopping baskets. It discusses the distinction between repeat and explore items in recommendations, introduces novel metrics for NBR evaluation, and highlights the variations in method performance across different datasets and scenarios, shedding light on the challenges and opportunities in this area of research.

The article's focus lies in examining the vulnerability of neural ranking models (NRMs) to adversarial attacks, particularly emphasizing the Word Substitution Ranking Attack (WSRA) task, aiming to enhance a target document's ranking by subtly altering its text. It covers topics such as proposing a novel PRADA method for generating adversarial examples, conducting experiments on web search datasets, and highlighting the importance of identifying NRM vulnerabilities to enhance model robustness.

The article primarily addresses the evaluation of click models used in search engines and their sensitivity to changes in ranking policies, known as policy distributional shift (PDS). It explores the impact of PDS on click models and proposes a new evaluation protocol to assess their robustness in various downstream tasks, such as click-through rate prediction and offline bandits, providing insights and guidelines for handling policy changes in click model deployments.

One of the key challenges in sequential recommendation is how to extract and represent user preferences. Traditional methods rely solely on predicting the next item. But user behavior may be driven by complex preferences. Therefore, these methods cannot make accurate recommendations when the available information user behavior is limited. To explore multiple user preferences, we propose a transformer-based sequential recommendation model, named MrTransformer (Multi-preference Transformer). For training MrTransformer, we devise a preference-editing-based self-supervised learning (SSL) mechanism that explores extra supervision signals based on relations with other sequences. The idea is to force the sequential recommendation model to discriminate between common and unique preferences in different sequences of interactions. By doing so, the sequential recommendation model is able to disentangle user preferences into multiple independent preference representations so as to improve user preference extraction and representation. We carry out extensive experiments on five benchmark datasets. MrTransformer with preference editing significantly outperforms state-of-the-art sequential recommendation methods in terms of Recall, MRR, and NDCG. We find that long sequences of interactions from which user preferences are harder to extract and represent benefit most from preference editing. [ABSTRACT FROM AUTHOR]

Content representation is a fundamental task in information retrieval. Representation learning is aimed at capturing features of an information object in a low-dimensional space. Most research on representation learning for heterogeneous information networks (HINs) focuses on static HINs. In practice, however, networks are dynamic and subject to constant change. In this article, we propose a novel and scalable representation learning model, M-DHIN, to explore the evolution of a dynamic HIN. We regard a dynamic HIN as a series of snapshots with different time stamps. We first use a static embedding method to learn the initial embeddings of a dynamic HIN at the first time stamp. We describe the features of the initial HIN via metagraphs, which retains more structural and semantic information than traditional path-oriented static models. We also adopt a complex embedding scheme to better distinguish between symmetric and asymmetric metagraphs. Unlike traditional models that process an entire network at each time stamp, we build a so-called change dataset that only includes nodes involved in a triadic closure or opening process, as well as newly added or deleted nodes. Then, we utilize the above metagraph-based mechanism to train on the change dataset. As a result of this setup, M-DHIN is scalable to large dynamic HINs since it only needs to model the entire HIN once while only the changed parts need to be processed over time. Existing dynamic embedding models only express the existing snapshots and cannot predict the future network structure. To equip M-DHIN with this ability, we introduce an LSTM-based deep autoencoder model that processes the evolution of the graph via an LSTM encoder and outputs the predicted graph. Finally, we evaluate the proposed model, M-DHIN, on real-life datasets and demonstrate that it significantly and consistently outperforms state-of-the-art models. [ABSTRACT FROM AUTHOR]

Network-based information has been widely explored and exploited in the information retrieval literature. Attributed networks, consisting of nodes, edges aswell as attributes describing properties of nodes, are a basic type of network-based data, and are especially useful for many applications. Examples include user profiling in social networks and item recommendation in user-item purchase networks. Learning useful and expressive representations of entities in attributed networks can provide more effective building blocks to down-stream network-based tasks such as link prediction and attribute inference. Practically, input features of attributed networks are normalized as unit directional vectors. However, most network embedding techniques ignore the spherical nature of inputs and focus on learning representations in a Gaussian or Euclidean space, which, we hypothesize, might lead to less effective representations. To obtain more effective representations of attributed networks, we investigate the problem of mapping an attributed networkwith unit normalized directional features into a non-Gaussian and non-Euclidean space. Specifically, we propose a hyperspherical variational co-embedding for attributed networks (HCAN), which is based on generalized variational auto-encoders for heterogeneous data with multiple types of entities. HCAN jointly learns latent embeddings for both nodes and attributes in a unified hyperspherical space such that the affinities between nodes and attributes can be captured effectively. We argue that this is a crucial feature in many real-world applications of attributed networks. Previous Gaussian network embedding algorithms break the assumption of uninformative prior, which leads to unstable results and poor performance. In contrast, HCAN embeds nodes and attributes as von Mises-Fisher distributions, and allows one to capture the uncertainty of the inferred representations. Experimental results on eight datasets show that HCAN yields better performance in a number of applications compared with nine state-of-the-art baselines. [ABSTRACT FROM AUTHOR]

A preliminary version of this article appeared in the proceedings of CIKM 2020 [10]. In this extension, we (1) propose another interest extractor, i.e., dynamic routing, in the implicit interest mining module, and another type of disagreement regularization, i.e., output disagreement regularization, in our interest-aware, diversity promoting loss; (2) investigate the performance of our multi-interest, diversified, sequential recommendation model with different interest extractors in implicit interest mining, i.e., multi-head attention vs. dynamic routing; (3) investigate the performance of multi-interest, diversified, sequential recommendation with various latent interests numbers; (4) explore the influence of the parameter λ on the performance of multi-interest, diversified, sequential recommendation; (5) investigate the performance of multiinterest, diversified, sequential recommendation with different types of disagreement regularization; (6) investigate the impact of maximum entropy regularization on the performance of multi-interest, diversified, sequential recommendation; (7) provide a case study to show the recommendations generated by multi-interest, diversified, sequential recommendation; (8) analyze the computational complexity of the baseline model as well as our proposal; and (9) survey more related work and conduct a more detailed analysis of the approach and experimental results. [ABSTRACT FROM AUTHOR]

In this article, we address the problem of answering complex information needs by conducting conversations with search engines, in the sense that users can express their queries in natural language and directly receive the information they need from a short system response in a conversationalmanner. Recently, there have been some attempts towards a similar goal, e.g., studies on Conversational Agents (CAs) and Conversational Search (CS). However, they either do not address complex information needs in search scenarios or they are limited to the development of conceptual frameworks and/or laboratory-based user studies. We pursue two goals in this article: (1) the creation of a suitable dataset, the Search as a Conversation (SaaC) dataset, for the development of pipelines for conversations with search engines, and (2) the development of a state-of-the-art pipeline for conversations with search engines, Conversations with Search Engines (CaSE), using this dataset. SaaC is built based on a multi-turn conversational search dataset, where we further employ workers from a crowdsourcing platform to summarize each relevant passage into a short, conversational response. CaSE enhances the state-of-the-art by introducing a supporting token identification module and a prior-aware pointer generator, which enables us to generate more accurate responses. We carry out experiments to show that CaSE is able to outperform strong baselines. We also conduct extensive analyses on the SaaC dataset to show where there is room for further improvement beyond CaSE. Finally, we release the SaaC dataset and the code for CaSE and all models used for comparison to facilitate future research on this topic. [ABSTRACT FROM AUTHOR]

Conversational search is a relatively young area of research that aims at automating an information-seeking dialogue. In this article, we help to position it with respect to other research areas within conversational artificial intelligence (AI) by analysing the structural properties of an information-seeking dialogue. To this end, we perform a large-scale dialogue analysis of more than 150K transcripts from 16 publicly available dialogue datasets. These datasets were collected to inform different dialogue-based tasks including conversational search. We extract different patterns of mixed initiative from these dialogue transcripts and use them to compare dialogues of different types. Moreover, we contrast the patterns found in information-seeking dialogues that are being used for research purposes with the patterns found in virtual reference interviews that were conducted by professional librarians. The insights we provide (1) establish close relations between conversational search and other conversational AI tasks and (2) uncover limitations of existing conversational datasets to inform future data collection tasks. [ABSTRACT FROM AUTHOR]

Contextual bandit problems are a natural fit for many information retrieval tasks, such as learning to rank, text classification, recommendation, and so on. However, existing learning methods for contextual bandit problems have one of two drawbacks: They either do not explore the space of all possible document rankings (i.e., actions) and, thus, may miss the optimal ranking, or they present suboptimal rankings to a user and, thus, may harm the user experience. We introduce a new learning method for contextual bandit problems, Safe Exploration Algorithm (SEA), which overcomes the above drawbacks. SEA starts by using a baseline (or production) ranking system (i.e., policy), which does not harm the user experience and, thus, is safe to execute but has suboptimal performance and, thus, needs to be improved. Then SEA uses counterfactual learning to learn a new policy based on the behavior of the baseline policy. SEA also uses high-confidence off-policy evaluation to estimate the performance of the newly learned policy. Once the performance of the newly learned policy is at least as good as the performance of the baseline policy, SEA starts using the new policy to execute new actions, allowing it to actively explore favorable regions of the action space. This way, SEA never performs worse than the baseline policy and, thus, does not harm the user experience, while still exploring the action space and, thus, being able to find an optimal policy. Our experiments using text classification and document retrieval confirm the above by comparing SEA (and a boundless variant called BSEA) to online and offline learning methods for contextual bandit problems. [ABSTRACT FROM AUTHOR]

We propose a Joint Neural Collaborative Filtering (J-NCF) method for recommender systems. The J-NCF model applies a joint neural network that couples deep feature learning and deep interaction modeling with a rating matrix. Deep feature learning extracts feature representations of users and items with a deep learning architecture based on a user-item rating matrix. Deep interaction modeling captures non-linear user-item interactions with a deep neural network using the feature representations generated by the deep feature learning process as input. J-NCF enables the deep feature learning and deep interaction modeling processes to optimize each other through joint training, which leads to improved recommendation performance. In addition, we design a new loss function for optimization that takes both implicit and explicit feedback, pointwise and pair-wise loss into account. Experiments on several real-world datasets show significant improvements of J-NCF over state-of-the-art methods, with improvements of up to 8.24% on the MovieLens 100K dataset, 10.81% on the MovieLens 1M dataset, and 10.21% on the Amazon Movies dataset in terms of HR@10. NDCG@10 improvements are 12.42%, 14.24%, and 15.06%, respectively. We also conduct experiments to evaluate the scalability and sensitivity of J-NCF. Our experiments show that the J-NCF model has a competitive recommendation performance with inactive users and different degrees of data sparsity when compared to state-of-the-art baselines. [ABSTRACT FROM AUTHOR]

Academic search engines have been widely used to access academic papers, where users' information needs are explicitly represented as search queries. Some modern recommender systems have taken one step further by predicting users' information needswithout the presence of an explicit query. In this article,we examine an academic paper recommender that sends out paper recommendations in email newsletters, based on the users' browsing history on the academic search engine. Specifically, we look at users who regularly browse papers on the search engine, and we sign up for the recommendation newsletters for the first time. We address the task of reranking the recommendation candidates that are generated by a production system for such users. We face the challenge that the users on whom we focus have not interacted with the recommender system before, which is a common scenario that every recommender system encounters when new users sign up. We propose an approach to reranking candidate recommendations that utilizes both paper content and user behavior. The approach is designed to suit the characteristics unique to our academic recommendation setting. For instance, content similarity measures can be used to find the closest match between candidate recommendations and the papers previously browsed by the user. To this end, we use a knowledge graph derived from paper metadata to compare entity similarities (papers, authors, and journals) in the embedding space. Since the users on whom we focus have no prior interactions with the recommender system, we propose a model to learn a mapping from users' browsed articles to user clicks on the recommendations. We combine both content and behavior into a hybrid reranking model that outperforms the production baseline significantly, providing a relative 13% increase in Mean Average Precision and 28% in Precision@1. Moreover, we provide a detailed analysis of the model components, highlighting where the performance boost comes from. The obtained insights reveal useful components for the reranking process and can be generalized to other academic recommendation settings as well, such as the utility of graph embedding similarity. Also, recent papers browsed by users provide stronger evidence for recommendation than historical ones. [ABSTRACT FROM AUTHOR]

Sentence regression is a type of extractive summarization that achieves state-of-the-art performance and is commonly used in practical systems. The most challenging task within the sentence regression framework is to identify discriminative features to represent each sentence. In this article, we study the use of sentence relations, e.g., Contextual Sentence Relations (CSR), Title Sentence Relations (TSR), and Query Sentence Relations (QSR), so as to improve the performance of sentence regression. CSR, TSR, and QSR refer to the relations between a main body sentence and its local context, its document title, and a given query, respectively. We propose a deep neural network model, Sentence Relation-based Summarization (SRSum), that consists of five sub-models, PriorSum, CSRSum, TSRSum, QSRSum, and SFSum. PriorSum encodes the latent semantic meaning of a sentence using a bi-gram convolutional neural network. SFSum encodes the surface information of a sentence, e.g., sentence length, sentence position, and so on. CSRSum, TSRSum, and QSRSum are three sentence relation sub-models corresponding to CSR, TSR, and QSR, respectively. CSRSum evaluates the ability of each sentence to summarize its local contexts. Specifically, CSRSum applies a CSR-based word-level and sentence-level attention mechanism to simulate the context-aware reading of a human reader, where words and sentences that have anaphoric relations or local summarization abilities are easily remembered and paid attention to. TSRSum evaluates the semantic closeness of each sentence with respect to its title, which usually reflects the main ideas of a document. TSRSum applies a TSR-based attentionmechanism to simulate people's reading abilitywith the main idea (title) in mind. QSRSum evaluates the relevance of each sentencewith given queries for the query-focused summarization. QSRSum applies a QSR-based attention mechanism to simulate the attentive reading of a human reader with some queries in mind. The mechanism can recognizewhich parts of the given queries are more likely answered by a sentence under consideration. Finally as a whole, SRSum automatically learns useful latent features by jointly learning representations of query sentences, content sentences, and title sentences as well as their relations. We conduct extensive experiments on six benchmark datasets, including generic multi-document summarization and query-focused multi-document summarization. On both tasks, SRSum achieves comparable or superior performance compared with state-of-the-art approaches in terms of multiple ROUGE metrics. [ABSTRACT FROM AUTHOR]

Query auto-completion assists web search users in formulating queries with a few keystrokes, helping them to avoid spelling mistakes and to produce clear query expressions, and so on. Previous work on query autocompletion mainly centers around returning a list of completions to users, aiming to push queries that are most likely intended by the user to the top positions but ignoring the redundancy among the query candidates in the list. Thus, semantically related queries matching the input prefix are often returned together. This may push valuable suggestions out of the list, given that only a limited number of candidates can be shown to the user, which may result in a less than optimal search experience. In this article, we consider the task of diversifying query auto-completion, which aims to return the correct query completions early in a ranked list of candidate completions and at the same time reduce the redundancy among query auto-completion candidates. We develop a greedy query selection approach that predicts query completions based on the current search popularity of candidate completions and on the aspects of previous queries in the same search session. The popularity of completion candidates at query time can be directly aggregated from query logs. However, query aspects are implicitly expressed by previous clicked documents in the search context. To determine the query aspect, we categorize clicked documents of a query using a hierarchy based on the open directory project. Bayesian probabilistic matrix factorization is applied to derive the distribution of queries over all aspects. We quantify the improvement of our greedy query selection model against a state-of-the-art baseline using two large-scale, real-world query logs and show that it beats the baseline in terms of well-known metrics used in query auto-completion and diversification. In addition, we conduct a side-by-side experiment to verify the effectiveness of our proposal. [ABSTRACT FROM AUTHOR]

A result page of a modern search engine often goes beyond a simple list of "10 blue links." Many specific user needs (e.g., News, Image, Video) are addressed by so-called aggregated or vertical search solutions: specially presented documents, often retrieved from specific sources, that stand out from the regular organic Web search results. When it comes to evaluating ranking systems, such complex result layouts raise their own challenges. This is especially true for so-called interleaving methods that have arisen as an important type of online evaluation: by mixing results from two different result pages, interleaving can easily break the desired Web layout in which vertical documents are grouped together, and hence hurt the user experience. We conduct an analysis of different interleaving methods as applied to aggregated search engine result pages. Apart from conventional interleaving methods, we propose two vertical-aware methods: one derived from the widely used Team-Draft Interleaving method by adjusting it in such a way that it respects vertical document groupings, and another based on the recently introduced Optimized Interleaving framework. We show that our proposed methods are better at preserving the user experience than existing interleaving methods while still performing well as a tool for comparing ranking systems. For evaluating our proposed vertical-aware interleaving methods, we use real-world click data as well as simulated clicks and simulated ranking systems. [ABSTRACT FROM AUTHOR]