论文检索
输入标题、作者或关键词，从 844 篇学术成果中精准定位
Vectors are used to represent unstructured data with their embeddings and associated attributes. Similarity search over large-scale vector datasets has gained significant interest from both industry and academia. It aims to identify the k nearest neighbors to a query object from vectors that satisfy a given attribute filter constraint. Despite its popularity, most solutions focus on single-sourced data and overlook the need for vector retrieval across federated datasets. To fill this gap, we introduce a new problem, federated vector similarity search with filters, which enables privacy-preserving vector retrieval over multi-sourced data held by mutually untrusted providers. While some solutions can be adapted, they struggle with low recall, excessive search latency, or high communication cost. To address these challenges, we propose FedVS, a privacy-preserving framework enhanced with indexing and pruning based on Trusted Execution Environment (TEE). We also provide a comprehensive theoretical analysis, including complexity, security, and approximation guarantees for recall. Moreover, we deploy our solution over real-world vector databases and conduct extensive experiments. The results demonstrate that our solution outperforms state-of-the-art methods in both effectiveness and efficiency.
Canonical algorithms for multi-armed bandits typically assume a stationary reward environment where the size of the action space (number of arms) is small. More recently developed methods typically relax only one of these assumptions: existing non-stationary bandit policies are designed for a small number of arms, while Lipschitz, linear, and Gaussian process bandit policies are designed to handle a large (or infinite) number of arms in stationary reward environments under constraints on the reward function. In this manuscript, we propose a novel policy to learn reward environments over a continuous space using Gaussian interpolation. We show that our method efficiently learns continuous Lipschitz reward functions with O*(√T) cumulative regret. Furthermore, our method naturally extends to non-stationary problems with a simple modification. We finally demonstrate that our method is computationally favorable (100-10000x faster) and experimentally outperforms sliding Gaussian process policies on datasets with non-stationarity and an extremely large number of arms.
We introduce a manifold analysis technique for neural network representations. Normalized Space Alignment (NSA) compares pairwise distances between two point clouds derived from the same source and having the same size, while potentially possessing differing dimensionalities. NSA can act as both an analytical tool and a differentiable loss function, providing a robust means of comparing and aligning representations across different layers and models. It satisfies the criteria necessary for both a dissimilarity metric and a neural network loss function. We showcase NSA's versatility by illustrating its utility as a representation space analysis metric and a structure-preserving loss function. NSA is not only computationally efficient, but it can also approximate the global structural discrepancy during mini-batching, facilitating its use in a wide variety of neural network training paradigms.
Revealing the underlying causal mechanisms in the real world is critical for scientific and technical progress. Despite advancements over the past decades, the lack of high-quality data and the inability of traditional causal discovery algorithms (TCDA) to fully comprehend the exact semantics of variables have long been major obstacles to the broader application of causal discovery. To address this issue, this paper proposes a novel causal modeling framework, LLM-CD, which integrates the metadata-based reasoning capabilities of large language models (LLMs) with the data-driven modeling abilities of TCDA for causal discovery. LLM-CD deeply couples the reasoning abilities of LLMs at various stages of TCDA, and enhances causal discovery through an iterative process. Due to the issues of overconfidence and hallucination in LLMs, LLM-CD quantifies and analyzes its uncertainty by incorporating evidence-based deep learning theory with the assumptions of TCDA. We utilize a large-scale de-identified real patient dataset provided by a hospital, a new dataset extracted from MIMIC-IV about the same disease (lung cancer), and two benchmark datasets to comprehensively evaluate LLM-CD. Extensive experimental results confirm the effectiveness and reliability of LLM-CD, with the highest improvement of 403.93% in the Recall and 25.77% in the Ratio metric across four datasets.
PERSCEN: Learning Personalized Interaction Pattern and Scenario Preference for Multi-Scenario Matching
PDF ↗With the expansion of business scales and scopes on online platforms, multi-scenario matching has become a mainstream solution to reduce maintenance costs and alleviate data sparsity. The key to effective multi-scenario recommendation lies in capturing both user preferences shared across all scenarios and scenario-aware preferences specific to each scenario. However, existing methods often overlook user-specific modeling, limiting the generation of personalized user representations. To address this, we propose PERSCEN, an innovative approach that incorporates user-specific modeling into multi-scenario matching. PERSCEN constructs a user-specific feature graph based on user characteristics and employs a lightweight graph neural network to capture higher-order interaction patterns, enabling personalized extraction of preferences shared across scenarios. Additionally, we leverage vector quantization techniques to distill scenario-aware preferences from users' behavior sequence within individual scenarios, facilitating user-specific and scenario-aware preference modeling. To enhance efficient and flexible information transfer, we introduce a progressive scenario-aware gated linear unit that allows fine-grained, low-latency fusion. Extensive experiments demonstrate that PERSCEN outperforms existing methods. Further efficiency analysis confirms that PERSCEN effectively balances performance with computational cost, ensuring its practicality for real-world industrial systems.
Machine learning models often suffer from performance degradation when dealing with class distributions that differ from the training distribution, a scenario commonly referred to as label shift. Addressing this challenge, this paper introduces Match, a novel adjustment approach that maximizes the likelihood of predicted probabilities under class prevalence constraints. Unlike existing methods such as retraining with instance re-weighting and the Bayes update rule, Match ensures that the adjusted class distribution aligns precisely with the prevalence estimates from quantifiers. By formulating the adjustment process as a binary integer linear optimization problem, Match benefits from efficient mixed-integer solvers. Extensive experiments demonstrate that Match outperforms the state-of-the-art in classifier adjustment with statistical significance, particularly in handling scenarios with imbalanced distributions.
Microservices architecture has become increasingly popular in modern software systems, yet its complexity also poses challenges for failure diagnosis. Existing literature has deficiencies in multimodal data mining, mainly in two aspects: first, the mining of multimodal data is insufficient, failing to fully exploit the rich information contained in different types of data; second, the analysis of spatiotemporal features of multimodal data is not thorough, failing to fully explore the potential associations of data in the temporal and spatial dimensions. To address these issues, this paper proposes a new method for anomaly detection in microservices. We transform three types of modal data into the wavelet domain and achieve fusion among modalities based on an attention mechanism, fully utilizing the inherent characteristics of multimodal data in the frequency and temporal domains. This fusion process can extract new information that does not exist in the original modalities, thereby enhancing the model's ability to detect anomalies. In addition, we propose a new diffusion probabilistic model (DDPM) based on spatiotemporal graphs, which combines spatiotemporal learning capabilities with the uncertainty measurement of DDPM to generate future samples in a non-autoregressive manner, achieving multi-horizon prediction. Experimental results show that our method significantly improves anomaly detection performance on three public datasets and a real-world production system, demonstrating its effectiveness in microservices anomaly detection.
Epidemiologists and social scientists have used the Network Scale-Up Method (NSUM) for over thirty years to estimate the size of a hidden sub-population within a social network. This method involves querying a subset of network nodes about the number of their neighbors belonging to the hidden sub-population. In general, NSUM assumes that the social network topology and the hidden sub-population distribution are well-behaved; hence, the NSUM estimate is close to the actual value. However, bounds on NSUM estimation errors have not been analytically proven. This paper provides analytical bounds on the error incurred by the two most popular NSUM estimators. These bounds assume that the queried nodes accurately provide their degree and the number of neighbors belonging to the hidden sub-population. Our key findings are twofold. First, we show that when an adversary designs the network and places the hidden sub-population, then the estimate can be a factor of Ω(√n) off from the real value (in a network with n nodes). Second, we also prove error bounds when the underlying network is randomly generated, showing that a small constant factor can be achieved with high probability using samples of logarithmic size O(log n). We present improved analytical bounds for Erdős-Rényi and Scale-Free networks. Our theoretical analysis is supported by an extensive set of numerical experiments designed to determine the effect of the sample size on the accuracy of the estimates in both synthetic and real networks.
Effective code optimization in compilers is crucial for computer and software engineering. The success of these optimizations primarily depends on the selection and ordering of the optimization passes applied to the code. While most compilers rely on fixed pass sequences, current methods to find the optimal sequence for specific programs either employ impractically slow search algorithms or learning methods that struggle to generalize to code unseen during training. To address these challenges, we introduce CompilerDream, the first world-model-based approach for general code optimization. CompilerDream features a compiler world model with a reward smoothing technique, enabling accurate simulation of optimization processes. Built on this model, code optimization agents can then be constructed via value prediction or direct optimization sequence generation. Trained on a large-scale program dataset, these agents serve as versatile code optimizers across diverse application scenarios and source-code languages. Our extensive experiments highlight CompilerDream's strong optimization capabilities for autotuning, where it leads the CompilerGym leaderboard. More importantly, the zero-shot generalization ability of large-scale trained compiler world model and agent, excels across diverse datasets, surpassing LLVM's built-in optimizations and state-of-the-art methods in both settings of value prediction and end-to-end code optimization.
Synthetic Electronic Health Record (EHR) time-series generation is crucial for advancing clinical machine learning models, as it helps address data scarcity by providing more training data. However, most existing approaches focus primarily on replicating statistical distributions and temporal dependencies of real-world data. We argue that fidelity to observed data alone does not guarantee better model performance, as common patterns may dominate, limiting the representation of rare but important conditions. This highlights the need for generate synthetic samples to improve performance of specific clinical models to fulfill their target outcomes. To address this, we propose TarDiff, a novel target-oriented diffusion framework that integrates task-specific influence guidance into the synthetic data generation process. Unlike conventional approaches that mimic training data distributions, TarDiff optimizes synthetic samples by quantifying their expected contribution to improving downstream model performance through influence functions. Specifically, we measure the reduction in task-specific loss induced by synthetic samples and embed this influence gradient into the reverse diffusion process, thereby steering the generation towards utility-optimized data. Evaluated on six publicly available EHR datasets, TarDiff achieves state-of-the-art performance, outperforming existing methods by up to 20.4% in AUPRC and 18.4% in AUROC. Our results demonstrate that TarDiff not only preserves temporal fidelity but also enhances downstream model performance, offering a robust solution to data scarcity and class imbalance in healthcare analytics.
Synthetic human trajectory data becoming increasingly prominent in various applications, including urban planning, traffic control, and crowd monitoring. Recent neural generative models for human trajectory data mostly follow an unconditional generative paradigm that relies on a pure data-driven imitative learning scheme, without considering the rich context of human mobility (e.g., social events or weather conditions) which may significantly impact the underlying human mobility patterns. Against this background, we propose Marionette, a Manipulatable generative model for human trajectory data with fine-grained conditions. Specifically, Marionette integrates both global and partial mobility-related contexts and extracts both sequence-level and event-level conditions. Afterward, it designs fine-grained and cascading conditioning mechanisms for modeling the temporal and spatial dynamics based on diffusion-alike Temporal Point Processes (TPPs) and discrete diffusion models, respectively, offering fine-grained controllable generative modeling of human trajectory data with both global and partial mobility-related contexts. We conduct a thorough evaluation on two real-world human trajectory datasets against a sizeable collection of baselines. Results show that our Marionette consistently outperforms the best baselines by 13.96-54.13% on statistical and distributional similarity metrics and by 9.36-40.63% in task-based data utility evaluation. Ablation studies verify our key design choices. Case studies also demonstrate the manipulability of Marionette in generating data in previously unseen scenarios.
Large language models (LLMs) have quickly emerged as practical and versatile tools that provide new solutions for a wide range of domains. In this paper, we consider the application of LLMs on symmetric tasks where a query is asked on an (unordered) bag of elements. Examples of such tasks include answering aggregate queries on a dataset. In general, when the bag contains a large number of elements, LLMs tend to overlook some elements, leading to challenges in generating accurate responses to the query. LLMs receive their inputs as ordered sequences. On the other hand, the input of a symmetric task is not ordered; hence reordering should not affect the LLM's response. Observing that LLMs are less likely to miss elements at certain positions of the input, we introduce the problem of LLM input reranking: to find a ranking of the input that maximizes the LLM's accuracy for the given query without making explicit assumptions about the query. Finding the optimal ranking requires identifying (i) the relevance of each input element for answering the query and (ii) the importance of each rank position for the LLM's attention. We develop algorithms for estimating these values efficiently utilizing a helper LLM. We conduct comprehensive experiments on different synthetic and real datasets to validate our proposal and to evaluate the effectiveness of our proposed algorithms. Our experiments confirm that our reranking approach improves the accuracy of the LLMs on symmetric tasks by up to 99% proximity to the optimum upper bound.
A key challenge impeding the widespread deployment of machine learning is overcoming the impact of statistical biases in the data. Models trained on unrepresentative data can perform worse than anticipated and differentially affect cross-sections of the population. Therefore, evaluating and vetting models based on an appropriate notion of fairness is often indispensable, making accurate estimation of fairness metrics a critical step to safeguard against deployment of unfair algorithms. It is often assumed that a fairness metric computed from the observed data is accurate. However, in presence of selection bias, also referred to as distributional shifts, fairness metric estimates too can have systematic application-specific errors. In this work we demonstrate this phenomenon and, relying on access to an unbiased unlabeled data, derive a semi-supervised approach to mitigate estimation errors emerging from the biased labeled data. Specifically, we introduce a novel selection bias model called ''sub-class-conditional invariance'' (SCC-invariance), that offers a flexible framework to effectively capture distributional shifts in the real-world data, particularly compared to traditional models such as label shift and covariate shift. Assuming a finite Gaussian mixture form for each class-conditional distribution, we then derive an Expectation-Maximization algorithm to estimate model parameters and correction weights necessary for computing unbiased estimates. We focus on three widely used fairness metrics--equal opportunity, predictive equality, and predictive parity--and demonstrate the effectiveness of our approach in improving their estimates on synthetic data. Finally, we apply our bias mitigation approach to clinical genetics and study the fairness of pathogenicity predictors across ancestral groups.
Co-branding has become a vital strategy for businesses aiming to expand market reach within recommendation systems. However, identifying effective cross-industry partnerships remains challenging due to resource imbalances, uncertain brand willingness, and ever-changing market conditions. In this paper, we provide the first systematic study of this problem and propose a unified online-offline framework to enable co-branding recommendations. Our approach begins by constructing a bipartite graph linking ''initiating'' and ''target'' brands to quantify co-branding probabilities and assess market benefits. During the online learning phase, we dynamically update the graph in response to market feedback, while striking a balance between exploring new collaborations for long-term gains and exploiting established partnerships for immediate benefits. To address the high initial co-branding costs, our framework mitigates redundant exploration, thereby enhancing short-term performance while ensuring sustainable strategic growth. In the offline optimization phase, our framework consolidates the interests of multiple sub-brands under the same parent brand to maximize overall returns, avoid excessive investment in single sub-brands, and reduce unnecessary costs associated with over-prioritizing a single sub-brand. We present a theoretical analysis of our approach, establishing a highly nontrivial sublinear regret bound for online learning in the complex co-branding problem, and enhancing the approximation guarantee for the NP-hard offline budget allocation optimization. Experiments on both synthetic and real-world co-branding datasets demonstrate the practical effectiveness of our framework, with at least 12% improvement.
Pretraining on Graph Neural Networks (GNNs) has shown great power in facilitating various downstream tasks. As pretraining generally requires huge amount of data and computational resources, the pretrained GNNs are high-value Intellectual Properties (IP) of the legitimate owner. However, adversaries may illegally copy and deploy the pretrained GNN models for their downstream tasks. Though initial efforts have been made to watermark GNN classifiers for IP protection, these methods are not applicable to self-supervised pretraining of GNN models. Hence, in this work, we propose a novel framework named PreGIP to watermark the pretraining of GNN encoder for IP protection while maintaining the high-quality of the embedding space. PreGIP incorporates a task-free watermarking loss to watermark the embedding space of pretrained GNN encoder. A finetuning-resistant watermark injection is further deployed. Theoretical analysis and extensive experiments show the effectiveness of PreGIP. The code can be find in https://anonymous.4open.science/r/PreGIP-semi/ and https://anonymous.4open.science/r/PreGIP-transfer/.
Medical imaging is crucial for diagnosing a patient's health condition, and accurate segmentation of these images is essential for isolating regions of interest to ensure precise diagnosis and treatment planning. Existing methods primarily rely on bounding boxes or point-based prompts, while few have explored text-related prompts, despite clinicians often describing their observations and instructions in natural language. To address this gap, we first propose a RAG-based free-form text prompt generator that leverages the domain corpus to generate diverse and realistic descriptions. Then, we introduce FLanS, a novel medical image segmentation model that handles various free-form text prompts, including professional anatomy-informed queries, anatomy-agnostic position-driven queries, and anatomy-agnostic size-driven queries. Additionally, our model also incorporates a symmetry-aware canonicalization module to ensure consistent, accurate segmentations across varying scan orientations and reduce confusion between the anatomical position of an organ and its appearance in the scan. FLanS is trained on a large-scale dataset of over 100k medical images from 7 public datasets. Comprehensive experiments demonstrate the model's superior language understanding and segmentation precision, along with a deep comprehension of the relationship between them, outperforming SOTA baselines on both in-domain and out-of-domain datasets.
Spatial crowdsourcing has become popular in recent years, but traditional tasks focus on one-to-one services with single skills like food delivery and ride-hailing. As societal needs grow more complex, there is a need for tasks requiring teams with multiple skills. Current team formation methods using workers from a single platform limit skill diversity, leading to potential task delays, lower quality, and revenue losses. Although cross-platform cooperation offers a potential solution to skill diversity limitations, it faces two challenges: (1) Data protection regulations mandate that platform's raw data must remain localized; (2) cross-platform cooperation incurs additional cooperation costs. To address these challenges, we first define the Cross-platform Online Team Formation (COTF) problem. We then propose a COTF framework and Random Cooperation Strategy to solve COTF problem. To enhance the effectiveness of cooperation, we further propose Precision Query Range Optimization Strategy (PQROS) for worker selection through adaptive range queries, and Dynamic Query Optimization (DQO) for cost-effective scheduling via predictive revenue modeling. Extensive experiments on real and synthetic datasets validate the effectiveness of our proposed methods.
Graph Contrastive Learning (GCL) emerged as a powerful paradigm in self-supervised graph representation learning. While earlier applications of GCL rely on homophily assumptions, spectral graph neural networks (GNNs) enhance the effectiveness of GCL on heterophilic graphs by incorporating both low-pass and high-pass filters. However, due to numerical considerations, existing approaches oversimplify low-pass and high-pass filters by modeling them as basic linear operations, failing to capture complex topological relationships. Here, we propose ArnoldiGCL, a novel algorithm that enables the application of complex spectral filters for Graph Contrastive Learning (GCL). Using Arnoldi orthonormalization-based Chebyshev interpolation, ArnoldiGCL overcomes the difficulties posed by ill-conditioned Vandermonde systems that arise in the modeling of complex filters. By introducing learnable filters, our method generates diverse spectral views and effectively captures nuanced graph structures. Theoretical analysis demonstrates that ArnoldiGCL accurately interpolates complex filters, thus forming a solid foundation for contrastive learning on graphs with complex structures. Extensive experiments on real-world datasets confirm that ArnoldiGCL significantly outperforms state-of-the-art GCL algorithms on both homophilic and heterophilic graphs, showcasing its robustness and versatility. The code for ArnoldiGCLis available as open source at: https://github.com/mustafaCoskunAgu/ArnoldiGCL
Large Language Models (LLMs) demonstrate significant persuasive capabilities in one-on-one interactions, but their influence within social networks, where interconnected users and complex opinion dynamics pose unique challenges, remains underexplored. This paper addresses the research question: Can LLMs generate meaningful content that maximizes user engagement on social networks? To answer this, we propose a pipeline using reinforcement learning with simulated feedback, where the network's response to LLM-generated content (i.e., the reward) is simulated through a formal engagement model. This approach bypasses the temporal cost and complexity of live experiments, enabling an efficient feedback loop between the LLM and the network under study. It also allows to control over endogenous factors such as the LLM's position within the social network and the distribution of opinions on a given topic. Our approach is adaptive to the opinion distribution of the underlying network and agnostic to the specifics of the engagement model, which is embedded as a plug-and-play component. Such flexibility makes it suitable for more complex engagement tasks and interventions in computational social science. Using our framework, we analyze the performance of LLMs in generating social engagement under different conditions, showcasing their full potential in this task. The experimental code is publicly available at https://github.com/mminici/Engagement-Driven-Content-Generation.