Quoc Viet Hung Nguyen

The rapid advancement of large language models (LLMs) has opened new possibilities for understanding users in search and recommendation. While traditional behavior-based or feature-driven user models rely primarily on explicit interactions or handcrafted representations, LLMs introduce a fundamentally different paradigm: LLM-powered user profiling, where user preferences, intents, and contextual attributes can be extracted, summarized, or reasoned about directly through natural language. This shift unlocks powerful new paths to achieve personalization but also raises pressing questions related to modeling fidelity, temporal dynamics, evaluation methodology, privacy, and responsible deployment. The LLM-UP workshop aims to bring together researchers and practitioners to systematize emerging progress in LLM-powered user profiling, identify open challenges, and explore opportunities for integrating such techniques into search and recommendation pipelines. The LLM-UP workshop adopts an interactive structure featuring lightning talks, panel discussions, and paper presentations to foster active engagement, cross-disciplinary dialogue, and community-driven agenda setting for this rapidly evolving field.

Time series analysis (TSA)–forecasting, anomaly detection, imputation, classification, and unified multi-task analytics– has become a battleground for sequence-modeling backbones, where Transformers, MLPs, convolutions, and state-space models compete on long context benchmarks. Since the 2023 release of Mamba, its linear time recurrence and input-dependent selectivity have triggered a surge of Mamba-based time-series models that report strong numbers yet resist like-for-like comparison. This survey provides the first focused treatment of Mamba for time series analysis, organized around four orthogonal perspectives: Model, Task, Data, and Application. The Model perspective formalizes a five-axis design space– tokenization, channel strategy, directional scan, hybridization, and decomposition– together with a three-pattern architectural toolbox of pure, bidirectional, and hybrid designs. The Task perspective re-indexes the corpus across the five canonical TSA tasks. The Data perspective re-tags it by data shape, spanning univariate, multivariate, spatio-temporal graph, and irregular series. The Application perspective then surveys deployments in healthcare, energy, traffic, climate, finance, activity recognition, and foundation-scale settings. Building on these views, we distill practical guidelines for variant selection, training recipes, and Mamba-specific pitfalls. We also catalog public implementations, datasets, and metrics, and we map out open frontiers in gain attribution, modeling regimes, and cross-task unification. An online repository is maintained at https://github.com/tamlhp/awesome-mamba-ts.

Zero-shot graph anomaly detection (GAD) has attracted increasing attention recent years, yet the heterogeneity of graph structures, features, and anomaly patterns across graphs make existing single GNN methods insufficiently expressive to model diverse anomaly mechanisms. In this regard, Mixture-of-experts (MoE) architectures provide a promising paradigm by integrating diverse GNN experts with complementary inductive biases, yet their effectiveness in zero-shot GAD is severely constrained by distribution shifts, leading to two key routing challenges. First, nodes often carry vastly different semantics across graphs, and straightforwardly performing routing based on their features is prone to generating biased or suboptimal expert assignments. Second, as anomalous graphs often exhibit pronounced distributional discrepancies, existing router designs fall short in capturing domain-invariant routing principles that generalize beyond the training graphs. To address these challenges, we propose a novel MoE framework with evolutionary router feature generation (EvoFG) for zero-shot GAD. To enhance MoE routing, we propose an evolutionary feature generation scheme that iteratively constructs and selects informative structural features via an LLM-based generator and Shapley-guided evaluation. Moreover, a memory-enhanced router with an invariant learning objective is designed to capture transferable routing patterns under distribution shifts. Extensive experiments on six benchmarks show that EvoFG consistently outperforms state-of-the-art baselines, achieving strong and stable zero-shot GAD performance.