arXiv subjects

Han Hu

Publications and source records attributed to Han Hu.

At least 127 records · Page 7Linked to original sources

arXiv2025

Merging Models on the Fly Without Retraining: A Sequential Approach to Scalable Continual Model Merging

Deep model merging represents an emerging research direction that combines multiple fine-tuned models to harness their specialized capabilities across different tasks and domains. Current model merging techniques focus on merging all available models simultaneously, with weight interpolation-based methods being the predominant approaches. However, these conventional approaches are not well-suited for scenarios where models become available sequentially, and they often suffer from high memory requirements and potential interference between tasks. In this study, we propose a training-free projection-based continual merging method that processes models sequentially through orthogonal projections of weight matrices and adaptive scaling mechanisms. Our method operates by projecting new parameter updates onto subspaces orthogonal to existing merged parameter updates while using an adaptive scaling mechanism to maintain stable parameter distances, enabling efficient sequential integration of task-specific knowledge. Our approach maintains constant memory complexity to the number of models, minimizes interference between tasks through orthogonal projections, and retains the performance of previously merged models through adaptive task vector scaling. Extensive experiments on CLIP-ViT models demonstrate that our method achieves a 5-8% average accuracy improvement while maintaining robust performance in different task orderings.

cs.LG↗

arXiv2025

LarvSeg: Exploring Image Classification Data For Large Vocabulary Semantic Segmentation via Category-wise Attentive Classifier

Scaling up the vocabulary of semantic segmentation models is extremely challenging because annotating large-scale mask labels is labour-intensive and time-consuming. Recently, language-guided segmentation models have been proposed to address this challenge. However, their performance drops significantly when applied to out-of-distribution categories. In this paper, we propose a new large vocabulary semantic segmentation framework, called LarvSeg. Different from previous works, LarvSeg leverages image classification data to scale the vocabulary of semantic segmentation models as large-vocabulary classification datasets usually contain balanced categories and are much easier to obtain. However, for classification tasks, the category is image-level, while for segmentation we need to predict the label at pixel level. To address this issue, we first propose a general baseline framework to incorporate image-level supervision into the training process of a pixel-level segmentation model, making the trained network perform semantic segmentation on newly introduced categories in the classification data. We then observe that a model trained on segmentation data can group pixel features of categories beyond the training vocabulary. Inspired by this finding, we design a category-wise attentive classifier to apply supervision to the precise regions of corresponding categories to improve the model performance. Extensive experiments demonstrate that LarvSeg significantly improves the large vocabulary semantic segmentation performance, especially in the categories without mask labels. For the first time, we provide a 21K-category semantic segmentation model with the help of ImageNet21K. The code is available at https://github.com/HaojunYu1998/large_voc_seg.

cs.CV↗

arXiv2024

Adaptive Channel Allocation for Robust Differentiable Architecture Search

Differentiable ARchiTecture Search (DARTS) has attracted much attention due to its simplicity and significant improvement in efficiency. However, the excessive accumulation of the skip connection, when training epochs become large, makes it suffer from weak stability and low robustness, thus limiting its practical applications. Many works have attempted to restrict the accumulation of skip connections by indicators or manual design. These methods, however, are susceptible to human priors and hyper-parameters. In this work, we suggest a more subtle and direct approach that no longer explicitly searches for skip connections in the search stage, based on the paradox that skip connections were proposed to guarantee the performance of very deep networks, but the networks in the search stage of differentiable architecture search are actually very shallow. Instead, by introducing channel importance ranking and channel allocation strategy, the skip connections are implicitly searched and automatically refilled unimportant channels in the evaluation stage. Our method, dubbed Adaptive Channel Allocation (ACA) strategy, is a general-purpose approach for differentiable architecture search, which universally works in DARTS variants without introducing human priors, indicators, or hyper-parameters. Extensive experiments on various datasets and DARTS variants verify that the ACA strategy is the most effective one among existing methods in improving robustness and dealing with the collapse issue when training epochs become large.

cs.CV↗

arXiv2024

ADEM-VL: Adaptive and Embedded Fusion for Efficient Vision-Language Tuning

Recent advancements in multimodal fusion have witnessed the remarkable success of vision-language (VL) models, which excel in various multimodal applications such as image captioning and visual question answering. However, building VL models requires substantial hardware resources, where efficiency is restricted by two key factors: the extended input sequence of the language model with vision features demands more computational operations, and a large number of additional learnable parameters increase memory complexity. These challenges significantly restrict the broader applicability of such models. To bridge this gap, we propose ADEM-VL, an efficient vision-language method that tunes VL models based on pretrained large language models (LLMs) by adopting a parameter-free cross-attention mechanism for similarity measurements in multimodal fusion. This approach only requires embedding vision features into the language space, significantly reducing the number of trainable parameters and accelerating both training and inference speeds. To enhance representation learning in fusion module, we introduce an efficient multiscale feature generation scheme that requires only a single forward pass through the vision encoder. Moreover, we propose an adaptive fusion scheme that dynamically discards less relevant visual information for each text token based on its attention score. This ensures that the fusion process prioritizes the most pertinent visual features. With experiments on various tasks including visual question answering, image captioning, and instruction-following, we demonstrate that our framework outperforms existing approaches. Specifically, our method surpasses existing methods by an average accuracy of 0.77% on ScienceQA dataset, with reduced training and inference latency, demonstrating the superiority of our framework. The code is available at https://github.com/Hao840/ADEM-VL.

cs.CV↗

arXiv2024

A Survey on Video Diffusion Models

The recent wave of AI-generated content (AIGC) has witnessed substantial success in computer vision, with the diffusion model playing a crucial role in this achievement. Due to their impressive generative capabilities, diffusion models are gradually superseding methods based on GANs and auto-regressive Transformers, demonstrating exceptional performance not only in image generation and editing, but also in the realm of video-related research. However, existing surveys mainly focus on diffusion models in the context of image generation, with few up-to-date reviews on their application in the video domain. To address this gap, this paper presents a comprehensive review of video diffusion models in the AIGC era. Specifically, we begin with a concise introduction to the fundamentals and evolution of diffusion models. Subsequently, we present an overview of research on diffusion models in the video domain, categorizing the work into three key areas: video generation, video editing, and other video understanding tasks. We conduct a thorough review of the literature in these three key areas, including further categorization and practical contributions in the field. Finally, we discuss the challenges faced by research in this domain and outline potential future developmental trends. A comprehensive list of video diffusion models studied in this survey is available at https://github.com/ChenHsing/Awesome-Video-Diffusion-Models.

cs.CV↗

arXiv2024

Joint Input and Output Coordination for Class-Incremental Learning

Incremental learning is nontrivial due to severe catastrophic forgetting. Although storing a small amount of data on old tasks during incremental learning is a feasible solution, current strategies still do not 1) adequately address the class bias problem, and 2) alleviate the mutual interference between new and old tasks, and 3) consider the problem of class bias within tasks. This motivates us to propose a joint input and output coordination (JIOC) mechanism to address these issues. This mechanism assigns different weights to different categories of data according to the gradient of the output score, and uses knowledge distillation (KD) to reduce the mutual interference between the outputs of old and new tasks. The proposed mechanism is general and flexible, and can be incorporated into different incremental learning approaches that use memory storage. Extensive experiments show that our mechanism can significantly improve their performance.

cs.LG↗

arXiv2024

SMILE: Zero-Shot Sparse Mixture of Low-Rank Experts Construction From Pre-Trained Foundation Models

Deep model training on extensive datasets is increasingly becoming cost-prohibitive, prompting the widespread adoption of deep model fusion techniques to leverage knowledge from pre-existing models. From simple weight averaging to more sophisticated methods like AdaMerging, model fusion effectively improves model performance and accelerates the development of new models. However, potential interference between parameters of individual models and the lack of interpretability in the fusion progress remain significant challenges. Existing methods often try to resolve the parameter interference issue by evaluating attributes of parameters, such as their magnitude or sign, or by parameter pruning. In this study, we begin by examining the fine-tuning of linear layers through the lens of subspace analysis and explicitly define parameter interference as an optimization problem to shed light on this subject. Subsequently, we introduce an innovative approach to model fusion called zero-shot Sparse MIxture of Low-rank Experts (SMILE) construction, which allows for the upscaling of source models into an MoE model without extra data or further training. Our approach relies on the observation that fine-tuning mostly keeps the important parts from the pre-training, but it uses less significant or unused areas to adapt to new tasks. Also, the issue of parameter interference, which is intrinsically intractable in the original parameter space, can be managed by expanding the dimensions. We conduct extensive experiments across diverse scenarios, such as image classification and text generation tasks, using full fine-tuning and LoRA fine-tuning, and we apply our method to large language models (CLIP models, Flan-T5 models, and Mistral-7B models), highlighting the adaptability and scalability of SMILE. Code is available at https://github.com/tanganke/fusion_bench

cs.LG↗

arXiv2024

Integrated Sensing, Communication, and Powering over Multi-antenna OFDM Systems

This paper considers a multi-functional orthogonal frequency division multiplexing (OFDM) system with integrated sensing, communication, and powering (ISCAP), in which a multi-antenna base station (BS) transmits OFDM signals to simultaneously deliver information to multiple information receivers (IRs), provide energy supply to multiple energy receivers (ERs), and sense potential targets based on the echo signals. To facilitate ISCAP, the BS employs the joint transmit beamforming design by sending dedicated sensing/energy beams jointly with information beams. Furthermore, we consider the beam scanning for sensing, in which the joint beams scan in different directions over time to sense potential targets. In order to ensure the sensing beam scanning performance and meet the communication and powering requirements, it is essential to properly schedule IRs and ERs and design the resource allocation over time, frequency, and space. More specifically, we optimize the joint transmit beamforming over multiple OFDM symbols and subcarriers, with the objective of minimizing the average beampattern matching error of beam scanning for sensing, subject to the constraints on the average communication rates at IRs and the average harvested power at ERs. We find converged high-quality solutions to the formulated problem by proposing efficient iterative algorithms based on advanced optimization techniques. We also develop various heuristic designs based on the principles of zero-forcing (ZF) beamforming, round-robin user scheduling, and time switching, respectively. Numerical results show that our proposed algorithms adaptively generate information and sensing/energy beams at each time-frequency slot to match the scheduled IRs/ERs with the desired scanning beam, significantly outperforming the heuristic designs.

eess.SP↗

arXiv2024

Sequential Federated Learning in Hierarchical Architecture on Non-IID Datasets

In a real federated learning (FL) system, communication overhead for passing model parameters between the clients and the parameter server (PS) is often a bottleneck. Hierarchical federated learning (HFL) that poses multiple edge servers (ESs) between clients and the PS can partially alleviate communication pressure but still needs the aggregation of model parameters from multiple ESs at the PS. To further reduce communication overhead, we bring sequential FL (SFL) into HFL for the first time, which removes the central PS and enables the model training to be completed only through passing the global model between two adjacent ESs for each iteration, and propose a novel algorithm adaptive to such a combinational framework, referred to as Fed-CHS. Convergence results are derived for strongly convex and non-convex loss functions under various data heterogeneity setups, which show comparable convergence performance with the algorithms for HFL or SFL solely. Experimental results provide evidence of the superiority of our proposed Fed-CHS on both communication overhead saving and test accuracy over baseline methods.

cs.LG↗

arXiv2024

ScalingFilter: Assessing Data Quality through Inverse Utilization of Scaling Laws

High-quality data is crucial for the pre-training performance of large language models. Unfortunately, existing quality filtering methods rely on a known high-quality dataset as reference, which can introduce potential bias and compromise diversity. In this paper, we propose ScalingFilter, a novel approach that evaluates text quality based on the perplexity difference between two language models trained on the same data, thereby eliminating the influence of the reference dataset in the filtering process. An theoretical analysis shows that ScalingFilter is equivalent to an inverse utilization of scaling laws. Through training models with 1.3B parameters on the same data source processed by various quality filters, we find ScalingFilter can improve zero-shot performance of pre-trained models in downstream tasks. To assess the bias introduced by quality filtering, we introduce semantic diversity, a metric of utilizing text embedding models for semantic representations. Extensive experiments reveal that semantic diversity is a reliable indicator of dataset diversity, and ScalingFilter achieves an optimal balance between downstream performance and semantic diversity.

cs.CL↗

arXiv2024

Multi-Granularity Hand Action Detection

Detecting hand actions in videos is crucial for understanding video content and has diverse real-world applications. Existing approaches often focus on whole-body actions or coarse-grained action categories, lacking fine-grained hand-action localization information. To fill this gap, we introduce the FHA-Kitchens (Fine-Grained Hand Actions in Kitchen Scenes) dataset, providing both coarse- and fine-grained hand action categories along with localization annotations. This dataset comprises 2,377 video clips and 30,047 frames, annotated with approximately 200k bounding boxes and 880 action categories. Evaluation of existing action detection methods on FHA-Kitchens reveals varying generalization capabilities across different granularities. To handle multi-granularity in hand actions, we propose MG-HAD, an End-to-End Multi-Granularity Hand Action Detection method. It incorporates two new designs: Multi-dimensional Action Queries and Coarse-Fine Contrastive Denoising. Extensive experiments demonstrate MG-HAD's effectiveness for multi-granularity hand action detection, highlighting the significance of FHA-Kitchens for future research and real-world applications. The dataset and source code are available at https://github.com/superZ678/MG-HAD.

cs.CV↗

arXiv2024

Secure Semantic Communications: From Perspective of Physical Layer Security

Semantic communications have been envisioned as a potential technique that goes beyond Shannon paradigm. Unlike modern communications that provide bit-level security, the eaves-dropping of semantic communications poses a significant risk of potentially exposing intention of legitimate user. To address this challenge, a novel deep neural network (DNN) enabled secure semantic communication (DeepSSC) system is developed by capitalizing on physical layer security. To balance the tradeoff between security and reliability, a two-phase training method for DNNs is devised. Particularly, Phase I aims at semantic recovery of legitimate user, while Phase II attempts to minimize the leakage of semantic information to eavesdroppers. The loss functions of DeepSSC in Phases I and II are respectively designed according to Shannon capacity and secure channel capacity, which are approximated with variational inference. Moreover, we define the metric of secure bilingual evaluation understudy (S-BLEU) to assess the security of semantic communications. Finally, simulation results demonstrate that DeepSSC achieves a significant boost to semantic security particularly in high signal-to-noise ratio regime, despite a minor degradation of reliability.

cs.IT↗

arXiv2024

Chat-like Asserts Prediction with the Support of Large Language Model

Unit testing is an essential component of software testing, with the assert statements playing an important role in determining whether the tested function operates as expected. Although research has explored automated test case generation, generating meaningful assert statements remains an ongoing challenge. While several studies have investigated assert statement generation in Java, limited work addresses this task in popular dynamically-typed programming languages like Python. In this paper, we introduce Chat-like execution-based Asserts Prediction (\tool), a novel Large Language Model-based approach for generating meaningful assert statements for Python projects. \tool utilizes the persona, Chain-of-Thought, and one-shot learning techniques in the prompt design, and conducts rounds of communication with LLM and Python interpreter to generate meaningful assert statements. We also present a Python assert statement dataset mined from GitHub. Our evaluation demonstrates that \tool achieves 64.7\% accuracy for single assert statement generation and 62\% for overall assert statement generation, outperforming the existing approaches. We also analyze the mismatched assert statements, which may still share the same functionality and discuss the potential help \tool could offer to the automated Python unit test generation. The findings indicate that \tool has the potential to benefit the SE community through more practical usage scenarios.

cs.SE↗

arXiv2024

A Near-Optimal Category Information Sampling in RFID Systems

In many RFID-enabled applications, objects are classified into different categories, and the information associated with each object's category (called category information) is written into the attached tag, allowing the reader to access it later. The category information sampling in such RFID systems, which is to randomly choose (sample) a few tags from each category and collect their category information, is fundamental for providing real-time monitoring and analysis in RFID. However, to the best of our knowledge, two technical challenges, i.e., how to guarantee a minimized execution time and reduce collection failure caused by missing tags, remain unsolved for this problem. In this paper, we address these two limitations by considering how to use the shortest possible time to sample a different number of random tags from each category and collect their category information sequentially in small batches. In particular, we first obtain a lower bound on the execution time of any protocol that can solve this problem. Then, we present a near-OPTimal Category information sampling protocol (OPT-C) that solves the problem with an execution time close to the lower bound. Finally, extensive simulation results demonstrate the superiority of OPT-C over existing protocols, while real-world experiments validate the practicality of OPT-C.

cs.NI↗

arXiv2024

Xwin-LM: Strong and Scalable Alignment Practice for LLMs

In this work, we present Xwin-LM, a comprehensive suite of alignment methodologies for large language models (LLMs). This suite encompasses several key techniques, including supervised finetuning (SFT), reward modeling (RM), rejection sampling finetuning (RS), and direct preference optimization (DPO). The key components are as follows: (1) Xwin-LM-SFT, models initially finetuned with high-quality instruction data; (2) Xwin-Pair, a large-scale, multi-turn preference dataset meticulously annotated using GPT-4; (3) Xwin-RM, reward models trained on Xwin-Pair, developed at scales of 7B, 13B, and 70B parameters; (4) Xwin-Set, a multiwise preference dataset in which each prompt is linked to 64 unique responses generated by Xwin-LM-SFT and scored by Xwin-RM; (5) Xwin-LM-RS, models finetuned with the highest-scoring responses from Xwin-Set; (6) Xwin-LM-DPO, models further optimized on Xwin-Set using the DPO algorithm. Our evaluations on AlpacaEval and MT-bench demonstrate consistent and significant improvements across the pipeline, demonstrating the strength and scalability of Xwin-LM. The repository https://github.com/Xwin-LM/Xwin-LM will be continually updated to foster community research.

cs.CL↗

arXiv2024

Enhancing GUI Exploration Coverage of Android Apps with Deep Link-Integrated Monkey

Mobile apps are ubiquitous in our daily lives for supporting different tasks such as reading and chatting. Despite the availability of many GUI testing tools, app testers still struggle with low testing code coverage due to tools frequently getting stuck in loops or overlooking activities with concealed entries. This results in a significant amount of testing time being spent on redundant and repetitive exploration of a few GUI pages. To address this, we utilize Android's deep links, which assist in triggering Android intents to lead users to specific pages and introduce a deep link-enhanced exploration method. This approach, integrated into the testing tool Monkey, gives rise to Delm (Deep Link-enhanced Monkey). Delm oversees the dynamic exploration process, guiding the tool out of meaningless testing loops to unexplored GUI pages. We provide a rigorous activity context mock-up approach for triggering existing Android intents to discover more activities with hidden entrances. We conduct experiments to evaluate Delm's effectiveness on activity context mock-up, activity coverage, method coverage, and crash detection. The findings reveal that Delm can mock up more complex activity contexts and significantly outperform state-of-the-art baselines with 27.2\% activity coverage, 21.13\% method coverage, and 23.81\% crash detection.

cs.SE↗

arXiv2024

Federated Learning with Only Positive Labels by Exploring Label Correlations

Federated learning aims to collaboratively learn a model by using the data from multiple users under privacy constraints. In this paper, we study the multi-label classification problem under the federated learning setting, where trivial solution and extremely poor performance may be obtained, especially when only positive data w.r.t. a single class label are provided for each client. This issue can be addressed by adding a specially designed regularizer on the server-side. Although effective sometimes, the label correlations are simply ignored and thus sub-optimal performance may be obtained. Besides, it is expensive and unsafe to exchange user's private embeddings between server and clients frequently, especially when training model in the contrastive way. To remedy these drawbacks, we propose a novel and generic method termed Federated Averaging by exploring Label Correlations (FedALC). Specifically, FedALC estimates the label correlations in the class embedding learning for different label pairs and utilizes it to improve the model training. To further improve the safety and also reduce the communication overhead, we propose a variant to learn fixed class embedding for each client, so that the server and clients only need to exchange class embeddings once. Extensive experiments on multiple popular datasets demonstrate that our FedALC can significantly outperform existing counterparts.

cs.LG↗

arXiv2024

Segment and Caption Anything

We propose a method to efficiently equip the Segment Anything Model (SAM) with the ability to generate regional captions. SAM presents strong generalizability to segment anything while is short for semantic understanding. By introducing a lightweight query-based feature mixer, we align the region-specific features with the embedding space of language models for later caption generation. As the number of trainable parameters is small (typically in the order of tens of millions), it costs less computation, less memory usage, and less communication bandwidth, resulting in both fast and scalable training. To address the scarcity problem of regional caption data, we propose to first pre-train our model on objection detection and segmentation tasks. We call this step weak supervision pretraining since the pre-training data only contains category names instead of full-sentence descriptions. The weak supervision pretraining allows us to leverage many publicly available object detection and segmentation datasets. We conduct extensive experiments to demonstrate the superiority of our method and validate each design choice. This work serves as a stepping stone towards scaling up regional captioning data and sheds light on exploring efficient ways to augment SAM with regional semantics. The project page, along with the associated code, can be accessed via https://xk-huang.github.io/segment-caption-anything/.

cs.CV↗