Text-video retrieval, a prominent sub-field within the domain of multimodal information retrieval, has witnessed remarkable growth in recent years. However, existing methods assume video scenes are consistent with unbiased descriptions. These limitations fail to align with real-world scenarios since descriptions can be influenced by annotator biases, diverse writing styles, and varying textual perspectives. To overcome the aforementioned problems, we introduce WAVER, a cross-domain knowledge distillation framework via vision-language models through open-vocabulary knowledge designed to tackle the challenge of handling different writing styles in video descriptions. WAVER capitalizes on the open-vocabulary properties that lie in pre-trained vision-language models and employs an implicit knowledge distillation approach to transfer text-based knowledge from a teacher model to a vision-based student. Empirical studies conducted across four standard benchmark datasets, encompassing various settings, provide compelling evidence that WAVER can achieve state-of-the-art performance in text-video retrieval task while handling writing-style variations.
Sound event localization and detection (SELD) is a task for the classification of sound events and the localization of direction of arrival (DoA) utilizing multichannel acoustic signals. Prior studies employ spectral and channel information as the embedding for temporal attention. However, this usage limits the deep neural network from extracting meaningful features from the spectral or spatial domains. Therefore, our investigation in this paper presents a novel framework termed the Channel-Spectro-Temporal Transformer (CST-former) that bolsters SELD performance through the independent application of attention mechanisms to distinct domains. The CST-former architecture employs distinct attention mechanisms to independently process channel, spectral, and temporal information. In addition, we propose an unfolded local embedding (ULE) technique for channel attention (CA) to generate informative embedding vectors including local spectral and temporal information. Empirical validation through experimentation on the 2022 and 2023 DCASE Challenge task3 datasets affirms the efficacy of employing attention mechanisms separated across each domain and the benefit of ULE, in enhancing SELD performance.
Time series domain adaptation stands as a pivotal and intricate challenge with diverse applications, including but not limited to human activity recognition, sleep stage classification, and machine fault diagnosis. Despite the numerous domain adaptation techniques proposed to tackle this complex problem, their primary focus has been on the common representations of time series data. This concentration might inadvertently lead to the oversight of valuable domain-specific information originating from different source domains. To bridge this gap, we introduce POND, a novel prompt-based deep learning model designed explicitly for multi-source time series domain adaptation. POND is tailored to address significant challenges, notably: 1) The unavailability of a quantitative relationship between meta-data information and time series distributions, and 2) The dearth of exploration into extracting domain-specific meta-data information. In this paper, we present an instance-level prompt generator and a fidelity loss mechanism to facilitate the faithful learning of meta-data information. Additionally, we propose a domain discrimination technique to discern domain-specific meta-data information from multiple source domains. Our approach involves a simple yet effective meta-learning algorithm to optimize the objective efficiently. Furthermore, we augment the model's performance by incorporating the Mixture of Expert (MoE) technique. The efficacy and robustness of our proposed POND model are extensively validated through experiments across 50 scenarios encompassing five datasets, which demonstrates that our proposed POND model outperforms the state-of-the-art methods by up to $66\%$ on the F1-score.
Despite the benefits of personalizing items and information tailored to users' needs, it has been found that recommender systems tend to introduce biases that favor popular items or certain categories of items, and dominant user groups. In this study, we aim to characterize the systematic errors of a recommendation system and how they manifest in various accountability issues, such as stereotypes, biases, and miscalibration. We propose a unified framework that distinguishes the sources of prediction errors into a set of key measures that quantify the various types of system-induced effects, both at the individual and collective levels. Based on our measuring framework, we examine the most widely adopted algorithms in the context of movie recommendation. Our research reveals three important findings: (1) Differences between algorithms: recommendations generated by simpler algorithms tend to be more stereotypical but less biased than those generated by more complex algorithms. (2) Disparate impact on groups and individuals: system-induced biases and stereotypes have a disproportionate effect on atypical users and minority groups (e.g., women and older users). (3) Mitigation opportunity: using structural equation modeling, we identify the interactions between user characteristics (typicality and diversity), system-induced effects, and miscalibration. We further investigate the possibility of mitigating system-induced effects by oversampling underrepresented groups and individuals, which was found to be effective in reducing stereotypes and improving recommendation quality. Our research is the first systematic examination of not only system-induced effects and miscalibration but also the stereotyping issue in recommender systems.
The quality of a face crop in an image is decided by many factors such as camera resolution, distance, and illumination condition. This makes the discrimination of face images with different qualities a challenging problem in realistic applications. However, most existing approaches are designed specifically for high-quality (HQ) or low-quality (LQ) images, and the performances would degrade for the mixed-quality images. Besides, many methods ask for pre-trained feature extractors or other auxiliary structures to support the training and the evaluation. In this paper, we point out that the key to better understand both the HQ and the LQ images simultaneously is to apply different learning methods according to their qualities. We propose a novel quality-guided joint training approach for mixed-quality face recognition, which could simultaneously learn the images of different qualities with a single encoder. Based on quality partition, classification-based method is employed for HQ data learning. Meanwhile, for the LQ images which lack identity information, we learn them with self-supervised image-image contrastive learning. To effectively catch up the model update and improve the discriminability of contrastive learning in our joint training scenario, we further propose a proxy-updated real-time queue to compose the contrastive pairs with features from the genuine encoder. Experiments on the low-quality datasets SCface and Tinyface, the mixed-quality dataset IJB-B, and five high-quality datasets demonstrate the effectiveness of our proposed approach in recognizing face images of different qualities.
Search engines are crucial as they provide an efficient and easy way to access vast amounts of information on the internet for diverse information needs. User queries, even with a specific need, can differ significantly. Prior research has explored the resilience of ranking models against typical query variations like paraphrasing, misspellings, and order changes. Yet, these works overlook how diverse demographics uniquely formulate identical queries. For instance, older individuals tend to construct queries more naturally and in varied order compared to other groups. This demographic diversity necessitates enhancing the adaptability of ranking models to diverse query formulations. To this end, in this paper, we propose a framework that integrates a novel rewriting pipeline that rewrites queries from various demographic perspectives and a novel framework to enhance ranking robustness. To be specific, we use Chain of Thought (CoT) technology to utilize Large Language Models (LLMs) as agents to emulate various demographic profiles, then use them for efficient query rewriting, and we innovate a robust Multi-gate Mixture of Experts (MMoE) architecture coupled with a hybrid loss function, collectively strengthening the ranking models' robustness. Our extensive experimentation on both public and industrial datasets assesses the efficacy of our query rewriting approach and the enhanced accuracy and robustness of the ranking model. The findings highlight the sophistication and effectiveness of our proposed model.
Future wireless multiple-input multiple-output (MIMO) communication systems will employ sub-6 GHz and millimeter wave (mmWave) frequency bands working cooperatively. Establishing a MIMO communication link usually relies on estimating channel state information (CSI) which is difficult to acquire at mmWave frequencies due to a low signal-to-noise ratio (SNR). In this paper, we propose three novel methods to estimate mmWave MIMO channels using out-of-band information obtained from the sub-6GHz band. We compare the proposed channel estimation methods with a conventional one utilizing only in-band information. Simulation results show that the proposed methods outperform the conventional mmWave channel estimation method in terms of achievable spectral efficiency, especially at low SNR and high K-factor.
Recently, multimodal prompting, which introduces learnable missing-aware prompts for all missing modality cases, has exhibited impressive performance. However, it encounters two critical issues: 1) The number of prompts grows exponentially as the number of modalities increases; and 2) It lacks robustness in scenarios with different missing modality settings between training and inference. In this paper, we propose a simple yet effective prompt design to address these challenges. Instead of using missing-aware prompts, we utilize prompts as modality-specific tokens, enabling them to capture the unique characteristics of each modality. Furthermore, our prompt design leverages orthogonality between prompts as a key element to learn distinct information across different modalities and promote diversity in the learned representations. Extensive experiments demonstrate that our prompt design enhances both performance and robustness while reducing the number of prompts.
Development of comprehensive prediction models are often of great interest in many disciplines of science, but datasets with information on all desired features typically have small sample sizes. In this article, we describe a transfer learning approach for building high-dimensional generalized linear models using data from a main study that has detailed information on all predictors, and from one or more external studies that have ascertained a more limited set of predictors. We propose using the external dataset(s) to build reduced model(s) and then transfer the information on underlying parameters for the analysis of the main study through a set of calibration equations, while accounting for the study-specific effects of certain design variables. We then use a generalized method of moment (GMM) with penalization for parameter estimation and develop highly scalable algorithms for fitting models taking advantage of the popular glmnet package. We further show that the use of adaptive-Lasso penalty leads to the oracle property of underlying parameter estimates and thus leads to convenient post-selection inference procedures. We conduct extensive simulation studies to investigate both predictive performance and post-selection inference properties of the proposed method. Finally, we illustrate a timely application of the proposed method for the development of risk prediction models for five common diseases using the UK Biobank study, combining baseline information from all study participants (500K) and recently released high-throughout proteomic data (# protein = 1500) on a subset (50K) of the participants.
Recommender systems can be privacy-sensitive. To protect users' private historical interactions, federated learning has been proposed in distributed learning for user representations. Using federated recommender (FedRec) systems, users can train a shared recommendation model on local devices and prevent raw data transmissions and collections. However, the recommendation model learned by a common FedRec may still be vulnerable to private information leakage risks, particularly attribute inference attacks, which means that the attacker can easily infer users' personal attributes from the learned model. Additionally, traditional FedRecs seldom consider the diverse privacy preference of users, leading to difficulties in balancing the recommendation utility and privacy preservation. Consequently, FedRecs may suffer from unnecessary recommendation performance loss due to over-protection and private information leakage simultaneously. In this work, we propose a novel user-consented federated recommendation system (UC-FedRec) to flexibly satisfy the different privacy needs of users by paying a minimum recommendation accuracy price. UC-FedRec allows users to self-define their privacy preferences to meet various demands and makes recommendations with user consent. Experiments conducted on different real-world datasets demonstrate that our framework is more efficient and flexible compared to baselines.