Deep Research agents driven by LLMs have automated the scholarly discovery pipeline, from planning and query formulation to iterative web exploration. Yet they remain constrained by a static, ''one-size-fits-all'' retrieval paradigm. Current systems fail to adaptively adjust the depth and breadth of exploration based on the user's existing expertise or latent interests, frequently resulting in reports that are either redundant for experts or overly dense for novices. To address this, we introduce Personalized Deep Research (PDR), a framework that integrates dynamic user context into the core retrieval-reasoning loop. Rather than treating personalization as a post-hoc formatting step, PDR unifies user profile modeling with iterative query development, dual-stage (private/public) retrieval, and context-aware synthesis. This allows the system to autonomously align research sub-goals with user intent and optimize the stopping criteria for evidence collection. To facilitate benchmarking, we release the PDR Dataset, covering four realistic user tasks, and propose a hybrid evaluation framework combining lexical metrics with LLM-based judgments to assess factual accuracy and personalization alignment. Experimental results against commercial baselines demonstrate that PDR significantly improves retrieval utility and report relevance, effectively bridging the gap between generic information retrieval and personalized knowledge acquisition. The resource is available to the public at~ https://github.com/Applied-Machine-Learning-Lab/SIGIR2026_PDR.

Cross-domain Sequential Recommendation (CDSR) has been proposed to enrich user-item interactions by incorporating information from various domains. Despite current progress, the domain imbalance issue and domain transition issue hinder further development of CDSR. The former presents a phenomenon where interactions in one domain dominate the entire behavior, leading to difficulty in capturing domain-specific features in the other domain. The latter points to the difficulty in capturing users' cross-domain preferences within the mixed interaction sequence, resulting in poor next-item prediction performance for specific domains. With world knowledge and powerful reasoning abilities, Large Language Models (LLMs) partially alleviate the above issues by functioning as both a generator and an encoder. However, current LLMs-enhanced CDSR methods are still under exploration, which fail to recognize the irrelevant noise and rough profiling problems. Thus, to address the aforementioned challenges, we propose an LLMs Enhanced Cross-domain Sequential Recommendation with Dual-phase Training (LLM-EDT). To address the domain imbalance issue while minimizing irrelevant noise, we propose the transferable item augmenter to adaptively generate possible cross-domain behaviors for users. Then, to alleviate the domain transition issue, we introduce a dual-phase training strategy to empower the domain-specific thread with a domain-shared background. As for the rough profiling problem, we devise a domain-aware profiling module to summarize the user's preference in each domain and adaptively aggregate them to generate comprehensive user profiles. The experiments on three public datasets validate the effectiveness of our proposed LLM-EDT. To ease reproducibility, we have released the detailed code online {https://github.com/Applied-Machine-Learning-Lab/SIGIR26_LLM-EDT}. © 2026 Copyright held by the owner/author(s).

Generative recommendation (GR) has shown great promise in industrial applications, particularly for candidate generation and end-to-end recommendations. However, existing GR training paradigms suffer from two fundamental mismatches with real-world deployment requirements. First, they optimize for point-wise prediction of a single ground-truth item, whereas practical systems must produce a diverse, high-value set of candidates. Second, they treat all user interactions as equally informative, ignoring their inherent differences in utility. Although reward-based fine-tuning offers a partial remedy, it often lacks token-level supervision. To address these challenges, we reformulate GR as a sequential set-generation problem and propose GFlowGR, a GFlowNet-based fine-tuning framework that explicitly aligns generation probabilities with item-level utilities. GFlowGR comprises three tightly integrated components, each addressing a key limitation of conventional fine-tuning: a trajectory sampler that constructs training trajectories from candidate sets to enable set-wise learning, a behavior-aware reward model that quantifies item utility to support value-aware optimization, and a GFlowNet objective that provides token-level supervision. Extensive experiments on three real-world datasets with two representative LLM-based GR backbones show consistent and significant improvements over strong baselines, validating the effectiveness of our approach. For real-world deployment, GFlowGR has been integrated into Taobao 's search advertising businesses, delivering a 0.4% relative improvement in annual revenue since its launch in mid-2025, corresponding to billion-level monetary gains. Code is available at https://github.com/Applied-Machine-Learning-Lab/SIGIR26_GFlowGR.

Generative Recommendation (GR), powered by Large Language Models (LLMs), represents a promising new paradigm for industrial recommender systems. However, their practical application is severely hindered by high inference latency, making them infeasible for high-throughput, real-time services and limiting their overall business impact. While Speculative Decoding (SD) has been proposed to accelerate the autoregressive generation process, existing implementations introduce new bottlenecks: they typically require separate draft models and model-based verifiers, which require additional training and increase latency overhead. In this paper, we address these challenges with NEZHA, a novel architecture that achieves hyperspeed decoding for GR systems without sacrificing recommendation quality. Specifically, NEZHA integrates a nimble autoregressive draft head directly into the primary model, enabling efficient self-drafting. This design, combined with a specialized input prompt structure, preserves the integrity of sequence-to-sequence generation. Furthermore, to tackle the critical problem of hallucination—a major source of performance degradation—we introduce an efficient, model-free verifier based on a hash set. We demonstrate the effectiveness of NEZHA through extensive experiments on public datasets and have successfully deployed the system on Taobao since October 2025, achieving 1.2% business improvement, translating to billion-level advertising revenue and serving hundreds of millions of daily active users. The code is available at https://github.com/Applied-Machine-Learning- Lab/WWW2026_NEZHA.

Multi-Scenario Recommendation (MSR) tasks, referring to building a unified model to enhance performance across all recommendation scenarios, have recently gained considerable attention. However, current research in MSR faces two significant challenges that hinder the field's development: the absence of uniform procedures for multi-scenario dataset processing, thus hindering fair comparisons, and most models being closed-source, which complicates comparisons with current SOTA models. Consequently, we introduce our benchmark, Scenario-Wise Rec, which comprises six public datasets and twelve baseline models, along with a training and evaluation pipeline. We further validate Scenario-Wise Rec on an industrial advertising dataset, underscoring its robustness. We hope the benchmark will give researchers clear insights into prior work, enabling them to develop novel models and thereby fostering a collaborative research ecosystem in MSR. Our source code is publicly available (https://github.com/Applied-Machine-Learning-Lab/Scenario-Wise-Rec).

Large Language Models (LLMs) have exhibited significant promise in recommender systems by empowering user profiles with their extensive world knowledge and superior reasoning capabilities. However, LLMs face challenges like unstable instruction compliance, modality gaps, and high inference latency, leading to textual noise and limiting their effectiveness in recommender systems. To address these challenges, we propose UserIP-Tuning, which uses prompt-tuning to infer user profiles. It integrates the causal relationship between user profiles and behavior sequences into LLMs' prompts. It employs Expectation Maximization (EM) to infer the embedded latent profile, minimizing textual noise by fixing the prompt template. Furthermore, a profile quantization codebook bridges the modality gap by categorizing profile embeddings into collaborative IDs pre-stored for online deployment. This improves time efficiency and reduces memory usage. Experiments show that UserIP-Tuning outperforms state-of-the-art recommendation algorithms. An industry application confirms its effectiveness, robustness, and transferability. The presented solution has been deployed in Huawei AppGallery's Explore page since May 2025, serving 2 million daily active users, delivering significant improvements in real-world recommendation scenarios. The code is publicly available for replication at https://github.com/Applied-Machine-Learning-Lab/UserIP-Tuning.

Due to exceptional reasoning and understanding abilities, the Large Language Model (LLM) has revolutionized the pattern of many fields, including recommender systems (RS). There has been a handful of research that focuses on empowering the RS by LLM. Recently, considering the latency and memory costs in real-world applications, LLM-Enhanced RS (LLMERS) is highlighted. This direction pushes the LLM into the online system with a large step by eliminating the utilization of LLM during inference. As a cutting-edge field, there is a clear need for a comprehensive survey to summarize this direction. In this survey, we systematically investigate the most up-to-date works of LLM-enhanced RS to boost this direction. Based on the component of an RS model that the LLM aims to augment, the basic taxonomy includes Knowledge Enhancement, Interaction Enhancement and Model Enhancement. Additionally, we identify several promising research directions. To facilitate access to the surveyed papers, we release a repository.

Cross-domain Sequential Recommendation (CDSR) aims to extract the preference from the user's historical interactions across various domains. Despite some progress in CDSR, two problems set the barrier for further advancements, i.e., overlap dilemma and transition complexity. The former means existing CDSR methods severely rely on users who own interactions on all domains to learn cross-domain item relationships, compromising the practicability. The latter refers to the difficulties in learning the complex transition patterns from the mixed behavior sequences. With powerful representation and reasoning abilities, Large Language Models (LLMs) are promising to address these two problems by bridging the items and capturing the user's preferences from a semantic view. Therefore, we propose an LLMs Enhanced Cross-domain Sequential Recommendation model (LLM4CDSR). To obtain the semantic item relationships, we first propose an LLM-based unified representation module to represent items. Then, a trainable adapter with contrastive regularization is designed to adapt the CDSR task. Besides, a hierarchical LLMs profiling module is designed to summarize user cross-domain preferences. Finally, these two modules are integrated into the proposed tri-thread framework to derive recommendations. We have conducted extensive experiments on three public cross-domain datasets, validating the effectiveness of LLM4CDSR. We have released the code online.

Auto-bidding, with its strong capability to optimize bidding decisions within dynamic and competitive online environments, has become a pivotal strategy for advertising platforms. Existing approaches typically employ rule-based strategies or Reinforcement Learning (RL) techniques. However, rule-based strategies lack the flexibility to adapt to time-varying market conditions, and RL-based methods struggle to capture essential historical dependencies and observations within Markov Decision Process (MDP) frameworks. Furthermore, these approaches often face challenges in ensuring strategy adaptability across diverse advertising objectives. Additionally, as offline training methods are increasingly adopted to facilitate the deployment and maintenance of stable online strategies, the issues of documented behavioral patterns and behavioral collapse resulting from training on fixed offline datasets become increasingly significant. To address these limitations, this paper introduces a novel offline Generative Auto-bidding framework with Value-Guided Explorations (GAVE). GAVE accommodates various advertising objectives through a score-based Return-To-Go (RTG) module. Moreover, GAVE integrates an action exploration mechanism with an RTG-based evaluation method to explore novel actions while ensuring stability-preserving updates. A learnable value function is also designed to guide the direction of action exploration and mitigate Out-of-Distribution (OOD) problems. Experimental results on two offline datasets and real-world deployments demonstrate that GAVE outperforms state-of-the-art baselines in both offline evaluations and online A/B tests. By applying the core methods of this framework, we proudly secured first place in the NeurIPS 2024 competition, 'AIGB Track: Learning Auto-Bidding Agents with Generative Models'.

Delivering superior search services is crucial for enhancing customer experience and driving revenue growth in e-commerce. Conventionally, search systems model user behaviors by combining user preference and query-item relevance statically, often through a fixed logical 'and' relationship. This paper reexamines existing approaches through a unified lens using causal graphs and Venn diagrams, uncovering two prevalent yet significant issues: entangled preference and relevance effects, and a collapsed modeling space. To surmount these challenges, our research introduces a novel framework, DRP, which enhances search accuracy through two components to reconstruct the behavior modeling space. Specifically, we implement preference editing to proactively remove the relevance effect from preference predictions, yielding untainted user preferences. Additionally, we employ adaptive fusion, which dynamically adjusts fusion criteria to align with the varying patterns of relevance and preference, facilitating more nuanced and tailored behavior predictions within the reconstructed modeling space. Empirical validation on two public datasets and a proprietary e-commerce search dataset underscores the superiority of our proposed methodology, demonstrating marked improvements in performance over existing approaches. The code is available at https://github.com/Applied-Machine-Learning-Lab/DRP.

Deep Recommender Systems (DRS) are essential for navigating the extensive data across various platforms in today's digital landscape. Current DRS models often treat all features equally and implement complex structures to enhance the capture of feature interactions. However, they may fail to recognize crucial user patterns due to not fully utilizing user-specific features for user modeling. Moreover, prevailing user modeling techniques concentrate exclusively on either the group or individual level, overlooking the potential insights from the unaddressed one. This oversight can miss shared group preferences or learn group patterns that conflict with individual preferences. To overcome these limitations, we introduce GPRec, a novel bi-level user modeling approach that substantially improves DRS. GPRec explicitly categorizes users into groups in a learnable manner and aligns them with corresponding group embeddings. We design the dual group embedding space to offer a diverse perspective on group preferences by contrasting positive and negative patterns. On the individual level, GPRec identifies personal preferences from ID-like features and refines the obtained individual representations to be independent of group ones, thereby providing a robust complement to the group-level modeling. We also present various strategies for the flexible integration of GPRec into various DRS models. Rigorous testing of GPRec on three public datasets has demonstrated significant improvements in recommendation quality. Additional experiments further explore crucial components of GPRec, its parameter sensitivity, and the group diversity. The implementation code is readily available online to facilitate future research and practical deployment: https://github.com/Applied-Machine-Learning-Lab/GPRec.

Deep Recommender Systems (DRS) are increasingly dependent on a large number of feature fields for more precise recommendations. Effective feature selection methods are consequently becoming critical for further enhancing the accuracy and optimizing storage efficiencies to align with the deployment demands. This research area, particularly in the context of DRS, is nascent and faces three core challenges. Firstly, variant experimental setups across research papers often yield unfair comparisons, obscuring practical insights. Secondly, the existing literature's lack of detailed analysis on selection attributes, based on large-scale datasets and a thorough comparison among selection techniques and DRS backbones, restricts the generalizability of findings and impedes deployment on DRS. Lastly, research often focuses on comparing the peak performance achievable by feature selection methods. This approach is typically computationally infeasible for identifying the optimal hyperparameters and overlooks evaluating the robustness and stability of these methods. To bridge these gaps, this paper presents ERASE, a comprehensive bEnchmaRk for feAture SElection for DRS. ERASE comprises a thorough evaluation of eleven feature selection methods, covering both traditional and deep learning approaches, across four public datasets, private industrial datasets, and a real-world commercial platform, achieving significant enhancement. Our code is available online for ease of reproduction.

As a representative information retrieval task, site recommendation, which aims at predicting the optimal sites for a brand or an institution to open new branches in an automatic data-driven way, is beneficial and crucial for brand development in modern business. However, there is no publicly available dataset so far and most existing approaches are limited to an extremely small scope of brands, which seriously hinders the research on site recommendation. Therefore, we collect, construct and release an open comprehensive dataset, namely OpenSiteRec, to facilitate and promote the research on site recommendation. Specifically, OpenSiteRec leverages a heterogeneous graph schema to represent various types of real-world entities and relations in four international metropolises. To evaluate the performance of the existing general methods on the site recommendation task, we conduct benchmarking experiments of several representative recommendation models on OpenSiteRec. Furthermore, we also highlight the potential application directions to demonstrate the wide applicability of OpenSiteRec. We believe that our OpenSiteRec dataset is significant and anticipated to encourage the development of advanced methods for site recommendation. OpenSiteRec is available online at https://OpenSiteRec.github.io/.

Multi-scenario recommender systems (MSRSs) have been increasingly used in real-world industrial platforms for their excellent advantages in mitigating data sparsity and reducing maintenance costs. However, conventional MSRSs usually use all relevant features indiscriminately and ignore that different kinds of features have varying importance under different scenarios, which may cause confusion and performance degradation. In addition, existing feature selection methods for deep recommender systems may lack the exploration of scenario relations. In this paper, we propose a novel automated multi-scenario feature selection (MultiFS) framework to bridge this gap, which is able to consider scenario relations and utilize a hierarchical gating mechanism to select features for each scenario. Specifically, MultiFS first efficiently obtains feature importance across all the scenarios through a scenario-shared gate. Then, some scenario-specific gate aims to identify feature importance to individual scenarios from a subset of the former with lower importance. Subsequently, MultiFS imposes constraints on the two gates to make the learning mechanism more feasible and combines the two to select exclusive features for different scenarios. We evaluate MultiFS and demonstrate its ability to enhance the multi-scenario model performance through experiments over two public multi-scenario benchmarks.

Deep recommender systems (DRS) are critical for current commercial online service providers, which address the issue of information overload by recommending items that are tailored to the user’s interests and preferences. They have unprecedented feature representations effectiveness and the capacity of modeling the non-linear relationships between users and items. Despite their advancements, DRS models, like other deep learning models, employ sophisticated neural network architectures and other vital components that are typically designed and tuned by human experts. This article will give a comprehensive summary of automated machine learning (AutoML) for developing DRS models. We first provide an overview of AutoML for DRS models and the related techniques. Then we discuss the state-of-the-art AutoML approaches that automate the feature selection, feature embeddings, feature interactions, and model training in DRS. We point out that the existing AutoML-based recommender systems are developing to a multi-component joint search with abstract search space and efficient search algorithm. Finally, we discuss appealing research directions and summarize the survey.

With the rapid growth of online medical platforms, more and more doctors are willing to manage and communicate with patients via online services. Considering the large volume and various patient conditions, identifying and classifying patients' medical records has become a crucial problem. To efficiently index these records, a common practice is to annotate them with semantically meaningful tags. However, manual labeling tags by doctors is impractical due to the possibility of thousands of tag candidates, which necessitates a tag recommender system. Due to the long tail distribution of tags and the dominance of low-activity doctors, as well as the unique uploaded medical records, this task is rather challenging. This paper proposes an efficient doctor specific tag recommendation framework for improved medical record management without side information. Specifically, we first utilize effective language models to learn the text representation. Then, we construct a doctor embedding learning module to enhance the recommendation quality by integrating implicit information within text representations and considering latent tag correlations to make more accurate predictions. Extensive experiment results demonstrate the effectiveness of our framework from the viewpoints of all doctors (20% improvement) or low-activity doctors (10% improvement).

With the rapid evolution of transformer architectures, researchers are exploring their application in sequential recommender systems (SRSs) and presenting promising performance on SRS tasks compared with former SRS models. However, most existing transformer-based SRS frameworks retain the vanilla attention mechanism, which calculates the attention scores between all item-item pairs. With this setting, redundant item interactions can harm the model performance and consume much computation time and memory. In this paper, we identify the sparse attention phenomenon in transformer-based SRS models and propose Sparse Transformer for sequential Recommendation tasks (STRec) to achieve the efficient computation and improved performance. Specifically, we replace self-attention with cross-attention, making the model concentrate on the most relevant item interactions. To determine these necessary interactions, we design a novel sampling strategy to detect relevant items based on temporal information. Extensive experimental results validate the effectiveness of STRec, which achieves the state-of-the-art accuracy while reducing 54% inference time and 70% memory cost. We also provide massive extended experiments to further investigate the property of our framework.

Multi-task Recommender Systems (MTRSs) has become increasingly prevalent in a variety of real-world applications due to their exceptional training efficiency and recommendation quality. However, conventional MTRSs often input all relevant feature fields without distinguishing their contributions to different tasks, which can lead to confusion and a decline in performance. Existing feature selection methods may neglect task relations or require significant computation during model training in multi-task setting. To this end, this paper proposes a novel Single-shot Feature Selection framework for MTRSs, referred to as MultiSFS, which is capable of selecting feature fields for each task while considering task relations in a single-shot manner. Specifically, MultiSFS first efficiently obtains task-specific feature importance through a single forward-backward pass. Then, a data-task bipartite graph is constructed to learn field-level task relations. Subsequently, MultiSFS merges the feature importance according to task relations and selects feature fields for different tasks. To demonstrate the effectiveness and properties of MultiSFS, we integrate it with representative MTRS models and evaluate on three real-world datasets. The implementation code is available online to ease reproducibility.

Recommender systems have become increasingly important in our daily lives since they play an important role in mitigating the information overload problem, especially in many user-oriented online services. Recommender systems aim to identify a set of items that best match users' explicit or implicit preferences, by utilizing the user and item interactions to improve the accuracy. With the fast advancement of deep neural networks (DNNs) in the past few decades, recommendation techniques have achieved promising performance. However, we still meet three inherent challenges to design deep recommender systems (DRS): 1) the majority of existing DRS are developed based on hand-crafted components, which requires ample expert knowledge recommender systems; 2) human error and bias can lead to suboptimal components, which reduces the recommendation effectiveness; 3) non-trivial time and engineering efforts are usually required to design the task-specific components in different recommendation scenarios.

Historical user-item interaction datasets are essential in training modern recommender systems for predicting user preferences. However, the arbitrary user behaviors in most recommendation scenarios lead to a large volume of noisy data instances being recorded, which cannot fully represent their true interests. While a large number of denoising studies are emerging in the recommender system community, all of them suffer from highly dynamic data distributions. In this paper, we propose a Deep Reinforcement Learning (DRL) based framework, AutoDenoise, with an Instance Denoising Policy Network, for denoising data instances with an instance selection manner in deep recommender systems. To be specific, AutoDenoise serves as an agent in DRL to adaptively select noise-free and predictive data instances, which can then be utilized directly in training representative recommendation models. In addition, we design an alternate two-phase optimization strategy to train and validate the AutoDenoise properly. In the searching phase, we aim to train the policy network with the capacity of instance denoising; in the validation phase, we find out and evaluate the denoised subset of data instances selected by the trained policy network, so as to validate its denoising ability. We conduct extensive experiments to validate the effectiveness of AutoDenoise combined with multiple representative recommender system models.

Feature selection plays an impactful role in deep recommender systems, which selects a subset of the most predictive features, so as to boost the recommendation performance and accelerate model optimization. The majority of existing feature selection methods, however, aim to select only a fixed subset of features. This setting cannot fit the dynamic and complex environments of practical recommender systems, where the contribution of a specific feature varies significantly across user-item interactions. In this paper, we propose an adaptive feature selection framework, AdaFS, for deep recommender systems. To be specific, we develop a novel controller network to automatically select the most relevant features from the whole feature space, which fits the dynamic recommendation environment better. Besides, different from classic feature selection approaches, the proposed controller can adaptively score each example of user-item interactions, and identify the most informative features correspondingly for subsequent recommendation tasks. We conduct extensive experiments based on two public benchmark datasets from a real-world recommender system. Experimental results demonstrate the effectiveness of AdaFS, and its excellent transferability to the most popular deep recommendation models.

Feature quality has an impactful effect on recommendation performance. Thereby, feature selection is a critical process in developing deep learning-based recommender systems. Most existing deep recommender systems, however, focus on designing sophisticated neural networks, while neglecting the feature selection process. Typically, they just feed all possible features into their proposed deep architectures, or select important features manually by human experts. The former leads to non-trivial embedding parameters and extra inference time, while the latter requires plenty of expert knowledge and human labor effort. In this work, we propose an AutoML framework that can adaptively select the essential feature fields in an automatic manner. Specifically, we first design a differentiable controller network, which is capable of automatically adjusting the probability of selecting a particular feature field; then, only selected feature fields are utilized to retrain the deep recommendation model. Extensive experiments on three benchmark datasets demonstrate the effectiveness of our framework. We conduct further experiments to investigate its properties, including the transferability, key components, and parameter sensitivity.