Abstract:Reinforcement Learning from Human/AI Feedback (RLHF/RLAIF) has been extensively utilized for preference alignment of text-to-image models. Existing methods face certain limitations in terms of both data and algorithm. For training data, most approaches rely on manual annotated preference data, either by directly fine-tuning the generators or by training reward models to provide training signals. However, the high annotation cost makes them difficult to scale up, the reward model consumes extra computation and cannot guarantee accuracy. From an algorithmic perspective, most methods neglect the value of text and only take the image feedback as a comparative signal, which is inefficient and sparse. To alleviate these drawbacks, we propose the InstructEngine framework. Regarding annotation cost, we first construct a taxonomy for text-to-image generation, then develop an automated data construction pipeline based on it. Leveraging advanced large multimodal models and human-defined rules, we generate 25K text-image preference pairs. Finally, we introduce cross-validation alignment method, which refines data efficiency by organizing semantically analogous samples into mutually comparable pairs. Evaluations on DrawBench demonstrate that InstructEngine improves SD v1.5 and SDXL's performance by 10.53% and 5.30%, outperforming state-of-the-art baselines, with ablation study confirming the benefits of InstructEngine's all components. A win rate of over 50% in human reviews also proves that InstructEngine better aligns with human preferences.
Abstract:Self-Supervised Video Hashing (SSVH) compresses videos into hash codes for efficient indexing and retrieval using unlabeled training videos. Existing approaches rely on random frame sampling to learn video features and treat all frames equally. This results in suboptimal hash codes, as it ignores frame-specific information density and reconstruction difficulty. To address this limitation, we propose a new framework, termed AutoSSVH, that employs adversarial frame sampling with hash-based contrastive learning. Our adversarial sampling strategy automatically identifies and selects challenging frames with richer information for reconstruction, enhancing encoding capability. Additionally, we introduce a hash component voting strategy and a point-to-set (P2Set) hash-based contrastive objective, which help capture complex inter-video semantic relationships in the Hamming space and improve the discriminability of learned hash codes. Extensive experiments demonstrate that AutoSSVH achieves superior retrieval efficacy and efficiency compared to state-of-the-art approaches. Code is available at https://github.com/EliSpectre/CVPR25-AutoSSVH.
Abstract:Video surveillance systems are crucial components for ensuring public safety and management in smart city. As a fundamental task in video surveillance, text-to-image person retrieval aims to retrieve the target person from an image gallery that best matches the given text description. Most existing text-to-image person retrieval methods are trained in a supervised manner that requires sufficient labeled data in the target domain. However, it is common in practice that only unlabeled data is available in the target domain due to the difficulty and cost of data annotation, which limits the generalization of existing methods in practical application scenarios. To address this issue, we propose a novel unsupervised domain adaptation method, termed Graph-Based Cross-Domain Knowledge Distillation (GCKD), to learn the cross-modal feature representation for text-to-image person retrieval in a cross-dataset scenario. The proposed GCKD method consists of two main components. Firstly, a graph-based multi-modal propagation module is designed to bridge the cross-domain correlation among the visual and textual samples. Secondly, a contrastive momentum knowledge distillation module is proposed to learn the cross-modal feature representation using the online knowledge distillation strategy. By jointly optimizing the two modules, the proposed method is able to achieve efficient performance for cross-dataset text-to-image person retrieval. acExtensive experiments on three publicly available text-to-image person retrieval datasets demonstrate the effectiveness of the proposed GCKD method, which consistently outperforms the state-of-the-art baselines.
Abstract:Self-supervised video hashing (SSVH) is a practical task in video indexing and retrieval. Although Transformers are predominant in SSVH for their impressive temporal modeling capabilities, they often suffer from computational and memory inefficiencies. Drawing inspiration from Mamba, an advanced state-space model, we explore its potential in SSVH to achieve a better balance between efficacy and efficiency. We introduce S5VH, a Mamba-based video hashing model with an improved self-supervised learning paradigm. Specifically, we design bidirectional Mamba layers for both the encoder and decoder, which are effective and efficient in capturing temporal relationships thanks to the data-dependent selective scanning mechanism with linear complexity. In our learning strategy, we transform global semantics in the feature space into semantically consistent and discriminative hash centers, followed by a center alignment loss as a global learning signal. Our self-local-global (SLG) paradigm significantly improves learning efficiency, leading to faster and better convergence. Extensive experiments demonstrate S5VH's improvements over state-of-the-art methods, superior transferability, and scalable advantages in inference efficiency. Code is available at https://github.com/gimpong/AAAI25-S5VH.
Abstract:Adaptation of pretrained vision-language models such as CLIP to various downstream tasks have raised great interest in recent researches. Previous works have proposed a variety of test-time adaptation (TTA) methods to achieve strong generalization without any knowledge of the target domain. However, existing training-required TTA approaches like TPT necessitate entropy minimization that involves large computational overhead, while training-free methods like TDA overlook the potential for information mining from the test samples themselves. In this paper, we break down the design of existing popular training-required and training-free TTA methods and bridge the gap between them within our framework. Specifically, we maintain a light-weight key-value memory for feature retrieval from instance-agnostic historical samples and instance-aware boosting samples. The historical samples are filtered from the testing data stream and serve to extract useful information from the target distribution, while the boosting samples are drawn from regional bootstrapping and capture the knowledge of the test sample itself. We theoretically justify the rationality behind our method and empirically verify its effectiveness on both the out-of-distribution and the cross-domain datasets, showcasing its applicability in real-world situations.
Abstract:Adaptation of pretrained vision-language models such as CLIP to various downstream tasks have raised great interest in recent researches. Previous works have proposed a variety of test-time adaptation (TTA) methods to achieve strong generalization without any knowledge of the target domain. However, existing training-required TTA approaches like TPT necessitate entropy minimization that involves large computational overhead, while training-free methods like TDA overlook the potential for information mining from the test samples themselves. In this paper, we break down the design of existing popular training-required and training-free TTA methods and bridge the gap between them within our framework. Specifically, we maintain a light-weight key-value memory for feature retrieval from instance-agnostic historical samples and instance-aware boosting samples. The historical samples are filtered from the testing data stream and serve to extract useful information from the target distribution, while the boosting samples are drawn from regional bootstrapping and capture the knowledge of the test sample itself. We theoretically justify the rationality behind our method and empirically verify its effectiveness on both the out-of-distribution and the cross-domain datasets, showcasing its applicability in real-world situations.
Abstract:With recent advances in large language models (LLMs), there has been emerging numbers of research in developing Semantic IDs based on LLMs to enhance the performance of recommendation systems. However, the dimension of these embeddings needs to match that of the ID embedding in recommendation, which is usually much smaller than the original length. Such dimension compression results in inevitable losses in discriminability and dimension robustness of the LLM embeddings, which motivates us to scale up the semantic representation. In this paper, we propose Mixture-of-Codes, which first constructs multiple independent codebooks for LLM representation in the indexing stage, and then utilizes the Semantic Representation along with a fusion module for the downstream recommendation stage. Extensive analysis and experiments demonstrate that our method achieves superior discriminability and dimension robustness scalability, leading to the best scale-up performance in recommendations.
Abstract:Cross-domain recommendation has attracted substantial interest in industrial apps such as Meituan, which serves multiple business domains via knowledge transfer and meets the diverse interests of users. However, existing methods typically follow an implicit modeling paradigm that blends the knowledge from both the source and target domains, and design intricate network structures to share learned embeddings or patterns between domains to improve recommendation accuracy. Since the transfer of interest signals is unsupervised, these implicit paradigms often struggle with the negative transfer resulting from differences in service functions and presentation forms across different domains. In this paper, we propose a simple and effective EXplicit Interest Transfer framework named EXIT to address the stated challenge. Specifically, we propose a novel label combination approach that enables the model to directly learn beneficial source domain interests through supervised learning, while excluding inappropriate interest signals. Moreover, we introduce a scene selector network to model the interest transfer intensity under fine-grained scenes. Offline experiments conducted on the industrial production dataset and online A/B tests validate the superiority and effectiveness of our proposed framework. Without complex network structures or training processes, EXIT can be easily deployed in the industrial recommendation system. EXIT has been successfully deployed in the online homepage recommendation system of Meituan App, serving the main traffic.
Abstract:Learned visual compression is an important and active task in multimedia. Existing approaches have explored various CNN- and Transformer-based designs to model content distribution and eliminate redundancy, where balancing efficacy (i.e., rate-distortion trade-off) and efficiency remains a challenge. Recently, state-space models (SSMs) have shown promise due to their long-range modeling capacity and efficiency. Inspired by this, we take the first step to explore SSMs for visual compression. We introduce MambaVC, a simple, strong and efficient compression network based on SSM. MambaVC develops a visual state space (VSS) block with a 2D selective scanning (2DSS) module as the nonlinear activation function after each downsampling, which helps to capture informative global contexts and enhances compression. On compression benchmark datasets, MambaVC achieves superior rate-distortion performance with lower computational and memory overheads. Specifically, it outperforms CNN and Transformer variants by 9.3% and 15.6% on Kodak, respectively, while reducing computation by 42% and 24%, and saving 12% and 71% of memory. MambaVC shows even greater improvements with high-resolution images, highlighting its potential and scalability in real-world applications. We also provide a comprehensive comparison of different network designs, underscoring MambaVC's advantages. Code is available at https://github.com/QinSY123/2024-MambaVC.
Abstract:Given a text query, partially relevant video retrieval (PRVR) aims to retrieve untrimmed videos containing relevant moments. Due to the lack of moment annotations, the uncertainty lying in clip modeling and text-clip correspondence leads to major challenges. Despite the great progress, existing solutions either sacrifice efficiency or efficacy to capture varying and uncertain video moments. What's worse, few methods have paid attention to the text-clip matching pattern under such uncertainty, exposing the risk of semantic collapse. To address these issues, we present GMMFormer v2, an uncertainty-aware framework for PRVR. For clip modeling, we improve a strong baseline GMMFormer with a novel temporal consolidation module upon multi-scale contextual features, which maintains efficiency and improves the perception for varying moments. To achieve uncertainty-aware text-clip matching, we upgrade the query diverse loss in GMMFormer to facilitate fine-grained uniformity and propose a novel optimal matching loss for fine-grained text-clip alignment. Their collaboration alleviates the semantic collapse phenomenon and neatly promotes accurate correspondence between texts and moments. We conduct extensive experiments and ablation studies on three PRVR benchmarks, demonstrating remarkable improvement of GMMFormer v2 compared to the past SOTA competitor and the versatility of uncertainty-aware text-clip matching for PRVR. Code is available at \url{https://github.com/huangmozhi9527/GMMFormer_v2}.