arXiv subjects

Chen Chen

Publications and source records attributed to Chen Chen.

At least 577 records · Page 32Linked to original sources

arXiv2024

Robust and Scalable Model Editing for Large Language Models

Large language models (LLMs) can make predictions using parametric knowledge--knowledge encoded in the model weights--or contextual knowledge--knowledge presented in the context. In many scenarios, a desirable behavior is that LLMs give precedence to contextual knowledge when it conflicts with the parametric knowledge, and fall back to using their parametric knowledge when the context is irrelevant. This enables updating and correcting the model's knowledge by in-context editing instead of retraining. Previous works have shown that LLMs are inclined to ignore contextual knowledge and fail to reliably fall back to parametric knowledge when presented with irrelevant context. In this work, we discover that, with proper prompting methods, instruction-finetuned LLMs can be highly controllable by contextual knowledge and robust to irrelevant context. Utilizing this feature, we propose EREN (Edit models by REading Notes) to improve the scalability and robustness of LLM editing. To better evaluate the robustness of model editors, we collect a new dataset, that contains irrelevant questions that are more challenging than the ones in existing datasets. Empirical results show that our method outperforms current state-of-the-art methods by a large margin. Unlike existing techniques, it can integrate knowledge from multiple edits, and correctly respond to syntactically similar but semantically unrelated inputs (and vice versa). The source code can be found at https://github.com/thunlp/EREN.

cs.CL↗

arXiv2024

E-Syn: E-Graph Rewriting with Technology-Aware Cost Functions for Logic Synthesis

Logic synthesis plays a crucial role in the digital design flow. It has a decisive influence on the final Quality of Results (QoR) of the circuit implementations. However, existing multi-level logic optimization algorithms often employ greedy approaches with a series of local optimization steps. Each step breaks the circuit into small pieces (e.g., k-feasible cuts) and applies incremental changes to individual pieces separately. These local optimization steps could limit the exploration space and may miss opportunities for significant improvements. To address the limitation, this paper proposes using e-graph in logic synthesis. The new workflow, named Esyn, makes use of the well-established e-graph infrastructure to efficiently perform logic rewriting. It explores a diverse set of equivalent Boolean representations while allowing technology-aware cost functions to better support delay-oriented and area-oriented logic synthesis. Experiments over a wide range of benchmark designs show our proposed logic optimization approach reaches a wider design space compared to the commonly used AIG-based logic synthesis flow. It achieves on average 15.29% delay saving in delay-oriented synthesis and 6.42% area saving for area-oriented synthesis.

cs.AR↗

arXiv2024

Large Language Model-informed ECG Dual Attention Network for Heart Failure Risk Prediction

Heart failure (HF) poses a significant public health challenge, with a rising global mortality rate. Early detection and prevention of HF could significantly reduce its impact. We introduce a novel methodology for predicting HF risk using 12-lead electrocardiograms (ECGs). We present a novel, lightweight dual-attention ECG network designed to capture complex ECG features essential for early HF risk prediction, despite the notable imbalance between low and high-risk groups. This network incorporates a cross-lead attention module and twelve lead-specific temporal attention modules, focusing on cross-lead interactions and each lead's local dynamics. To further alleviate model overfitting, we leverage a large language model (LLM) with a public ECG-Report dataset for pretraining on an ECG-report alignment task. The network is then fine-tuned for HF risk prediction using two specific cohorts from the UK Biobank study, focusing on patients with hypertension (UKB-HYP) and those who have had a myocardial infarction (UKB-MI).The results reveal that LLM-informed pre-training substantially enhances HF risk prediction in these cohorts. The dual-attention design not only improves interpretability but also predictive accuracy, outperforming existing competitive methods with C-index scores of 0.6349 for UKB-HYP and 0.5805 for UKB-MI. This demonstrates our method's potential in advancing HF risk assessment with clinical complex ECG data.

q-bio.QM↗

arXiv2024

Enhanced Few-Shot Class-Incremental Learning via Ensemble Models

Few-shot class-incremental learning (FSCIL) aims to continually fit new classes with limited training data, while maintaining the performance of previously learned classes. The main challenges are overfitting the rare new training samples and forgetting old classes. While catastrophic forgetting has been extensively studied, the overfitting problem has attracted less attention in FSCIL. To tackle overfitting challenge, we design a new ensemble model framework cooperated with data augmentation to boost generalization. In this way, the enhanced model works as a library storing abundant features to guarantee fast adaptation to downstream tasks. Specifically, the multi-input multi-output ensemble structure is applied with a spatial-aware data augmentation strategy, aiming at diversifying the feature extractor and alleviating overfitting in incremental sessions. Moreover, self-supervised learning is also integrated to further improve the model generalization. Comprehensive experimental results show that the proposed method can indeed mitigate the overfitting problem in FSCIL, and outperform the state-of-the-art methods.

cs.CV↗

arXiv2024

FedMef: Towards Memory-efficient Federated Dynamic Pruning

Federated learning (FL) promotes decentralized training while prioritizing data confidentiality. However, its application on resource-constrained devices is challenging due to the high demand for computation and memory resources to train deep learning models. Neural network pruning techniques, such as dynamic pruning, could enhance model efficiency, but directly adopting them in FL still poses substantial challenges, including post-pruning performance degradation, high activation memory usage, etc. To address these challenges, we propose FedMef, a novel and memory-efficient federated dynamic pruning framework. FedMef comprises two key components. First, we introduce the budget-aware extrusion that maintains pruning efficiency while preserving post-pruning performance by salvaging crucial information from parameters marked for pruning within a given budget. Second, we propose scaled activation pruning to effectively reduce activation memory footprints, which is particularly beneficial for deploying FL to memory-limited devices. Extensive experiments demonstrate the effectiveness of our proposed FedMef. In particular, it achieves a significant reduction of 28.5% in memory footprint compared to state-of-the-art methods while obtaining superior accuracy.

cs.LG↗

arXiv2024

A Dual-Augmentor Framework for Domain Generalization in 3D Human Pose Estimation

3D human pose data collected in controlled laboratory settings present challenges for pose estimators that generalize across diverse scenarios. To address this, domain generalization is employed. Current methodologies in domain generalization for 3D human pose estimation typically utilize adversarial training to generate synthetic poses for training. Nonetheless, these approaches exhibit several limitations. First, the lack of prior information about the target domain complicates the application of suitable augmentation through a single pose augmentor, affecting generalization on target domains. Moreover, adversarial training's discriminator tends to enforce similarity between source and synthesized poses, impeding the exploration of out-of-source distributions. Furthermore, the pose estimator's optimization is not exposed to domain shifts, limiting its overall generalization ability. To address these limitations, we propose a novel framework featuring two pose augmentors: the weak and the strong augmentors. Our framework employs differential strategies for generation and discrimination processes, facilitating the preservation of knowledge related to source poses and the exploration of out-of-source distributions without prior information about target poses. Besides, we leverage meta-optimization to simulate domain shifts in the optimization process of the pose estimator, thereby improving its generalization ability. Our proposed approach significantly outperforms existing methods, as demonstrated through comprehensive experiments on various benchmark datasets.Our code will be released at \url{https://github.com/davidpengucf/DAF-DG}.

cs.CV↗

arXiv2024

InsCL: A Data-efficient Continual Learning Paradigm for Fine-tuning Large Language Models with Instructions

Instruction tuning effectively optimizes Large Language Models (LLMs) for downstream tasks. Due to the changing environment in real-life applications, LLMs necessitate continual task-specific adaptation without catastrophic forgetting. Considering the heavy computational cost, replay-based Continual Learning (CL) methods are the simplest and most widely used for LLMs to address the forgetting issue. However, traditional replay-based methods do not fully utilize instructions to customize the replay strategy. In this work, we propose a novel paradigm called Instruction-based Continual Learning (InsCL). InsCL dynamically replays previous data based on task similarity, calculated by Wasserstein Distance with instructions. Moreover, we further introduce an Instruction Information Metric (InsInfo) to quantify the complexity and diversity of instructions. According to InsInfo, InsCL guides the replay process more inclined to high-quality data. We conduct extensive experiments over 16 tasks with different training orders, observing consistent performance improvements of InsCL. When all tasks have been trained, InsCL achieves performance gains of 3.0 Relative Gain compared with Random Replay, and 27.96 Relative Gain compared with No Replay.

cs.CL↗

arXiv2024

MOFI: Learning Image Representations from Noisy Entity Annotated Images

We present MOFI, Manifold OF Images, a new vision foundation model designed to learn image representations from noisy entity annotated images. MOFI differs from previous work in two key aspects: (i) pre-training data, and (ii) training recipe. Regarding data, we introduce a new approach to automatically assign entity labels to images from noisy image-text pairs. Our approach involves employing a named entity recognition model to extract entities from the alt-text, and then using a CLIP model to select the correct entities as labels of the paired image. It's a simple, cost-effective method that can scale to handle billions of web-mined image-text pairs. Through this method, we have created Image-to-Entities (I2E), a new dataset with 1 billion images and 2 million distinct entities, covering rich visual concepts in the wild. Building upon the I2E dataset, we study different training recipes like supervised pre-training, contrastive pre-training, and multi-task learning. For contrastive pre-training, we treat entity names as free-form text, and further enrich them with entity descriptions. Experiments show that supervised pre-training with large-scale fine-grained entity labels is highly effective for image retrieval tasks, and multi-task training further improves the performance. The final MOFI model achieves 86.66% mAP on the challenging GPR1200 dataset, surpassing the previous state-of-the-art performance of 72.19% from OpenAI's CLIP model. Further experiments on zero-shot and linear probe image classification also show that MOFI outperforms a CLIP model trained on the original image-text data, demonstrating the effectiveness of the I2E dataset in learning strong image representations. We release our code and model weights at https://github.com/apple/ml-mofi.

cs.CV↗

arXiv2024

Learning Semantic Proxies from Visual Prompts for Parameter-Efficient Fine-Tuning in Deep Metric Learning

Deep Metric Learning (DML) has long attracted the attention of the machine learning community as a key objective. Existing solutions concentrate on fine-tuning the pre-trained models on conventional image datasets. As a result of the success of recent pre-trained models trained from larger-scale datasets, it is challenging to adapt the model to the DML tasks in the local data domain while retaining the previously gained knowledge. In this paper, we investigate parameter-efficient methods for fine-tuning the pre-trained model for DML tasks. In particular, we propose a novel and effective framework based on learning Visual Prompts (VPT) in the pre-trained Vision Transformers (ViT). Based on the conventional proxy-based DML paradigm, we augment the proxy by incorporating the semantic information from the input image and the ViT, in which we optimize the visual prompts for each class. We demonstrate that our new approximations with semantic information are superior to representative capabilities, thereby improving metric learning performance. We conduct extensive experiments to demonstrate that our proposed framework is effective and efficient by evaluating popular DML benchmarks. In particular, we demonstrate that our fine-tuning method achieves comparable or even better performance than recent state-of-the-art full fine-tuning works of DML while tuning only a small percentage of total parameters.

cs.CV↗

arXiv2024

WhisperFuzz: White-Box Fuzzing for Detecting and Locating Timing Vulnerabilities in Processors

Timing vulnerabilities in processors have emerged as a potent threat. As processors are the foundation of any computing system, identifying these flaws is imperative. Recently fuzzing techniques, traditionally used for detecting software vulnerabilities, have shown promising results for uncovering vulnerabilities in large-scale hardware designs, such as processors. Researchers have adapted black-box or grey-box fuzzing to detect timing vulnerabilities in processors. However, they cannot identify the locations or root causes of these timing vulnerabilities, nor do they provide coverage feedback to enable the designer's confidence in the processor's security. To address the deficiencies of the existing fuzzers, we present WhisperFuzz--the first white-box fuzzer with static analysis--aiming to detect and locate timing vulnerabilities in processors and evaluate the coverage of microarchitectural timing behaviors. WhisperFuzz uses the fundamental nature of processors' timing behaviors, microarchitectural state transitions, to localize timing vulnerabilities. WhisperFuzz automatically extracts microarchitectural state transitions from a processor design at the register-transfer level (RTL) and instruments the design to monitor the state transitions as coverage. Moreover, WhisperFuzz measures the time a design-under-test (DUT) takes to process tests, identifying any minor, abnormal variations that may hint at a timing vulnerability. WhisperFuzz detects 12 new timing vulnerabilities across advanced open-sourced RISC-V processors: BOOM, Rocket Core, and CVA6. Eight of these violate the zero latency requirements of the Zkt extension and are considered serious security vulnerabilities. Moreover, WhisperFuzz also pinpoints the locations of the new and the existing vulnerabilities.

cs.CR↗

arXiv2024

How do Older Adults Set Up Voice Assistants? Lessons Learned from a Deployment Experience for Older Adults to Set Up Standalone Voice Assistants

While standalone Voice Assistants (VAs) are promising to support older adults' daily routine and wellbeing management, onboarding and setting up these devices can be challenging. Although some older adults choose to seek assistance from technicians and adult children, easy set up processes that facilitate independent use are still critical, especially for those who do not have access to external resources. We aim to understand the older adults' experience while setting up commercially available voice-only and voice-first screen-based VAs. Rooted in participants observations and semi-structured interviews, we designed a within-subject study with 10 older adults using Amazon Echo Dot and Echo Show. We identified the values of the built-in touchscreen and the instruction documents, as well as the impact of form factors, and outline important directions to support older adult independence with VAs.

cs.HC↗

arXiv2024

Bound-extended mode transition in type-II synthetic photonic Weyl heterostructures

Photonic structures with Weyl points (WPs), including type-I and type-II, promise nontrivial surface modes and intriguing light manipulations for their three-dimensional topological bands. While previous studies mainly focus on exploring WPs in a uniform Weyl structure, here we establish Weyl heterostructures (i.e., a nonuniform Weyl lattice) with different rotational orientations in the synthetic dimension by nanostructured photonic waveguides. In this work, we unveil a transition between bound and extended modes on the interface of type-II Weyl heterostructures by tuning their rotational phases, despite the reversed topological order across the interface. This mode transition is also manifested from the total transmission to total reflection at the interface. All of these unconventional effects are attributed to the tilted dispersion of type-II Weyl band structure that can lead to mismatched bands and gaps across the interface. As a comparison, the type-I Weyl heterostructures lack the phase transition due to the untilted band structure. This work establishes a flexible scheme of artificial Weyl heterostructures that opens a new avenue towards high-dimensional topological effects and significantly enhances our capabilities in on-chip light manipulations.

physics.optics↗

arXiv2024

Adversarial Attacks on Fairness of Graph Neural Networks

Fairness-aware graph neural networks (GNNs) have gained a surge of attention as they can reduce the bias of predictions on any demographic group (e.g., female) in graph-based applications. Although these methods greatly improve the algorithmic fairness of GNNs, the fairness can be easily corrupted by carefully designed adversarial attacks. In this paper, we investigate the problem of adversarial attacks on fairness of GNNs and propose G-FairAttack, a general framework for attacking various types of fairness-aware GNNs in terms of fairness with an unnoticeable effect on prediction utility. In addition, we propose a fast computation technique to reduce the time complexity of G-FairAttack. The experimental study demonstrates that G-FairAttack successfully corrupts the fairness of different types of GNNs while keeping the attack unnoticeable. Our study on fairness attacks sheds light on potential vulnerabilities in fairness-aware GNNs and guides further research on the robustness of GNNs in terms of fairness.

cs.LG↗

arXiv2024

Resilient Mobile Energy Storage Resources Based Distribution Network Restoration in Interdependent Power-Transportation-Information Networks

The interactions between power, transportation, and information networks (PTIN), are becoming more profound with the advent of smart city technologies. Existing mobile energy storage resource (MESR)-based power distribution network (PDN) restoration schemes often neglect the interdependencies among PTIN, thus, efficient PDN restoration cannot be achieved. This paper outlines the interacting factors of power supply demand, traffic operation efficiency, communication coverage, electric vehicle (EV) deployment capability, and PDN controllability among PTIN and further develops a PTIN-interacting model to reflect the chained recovery effect of the MESR-based restoration process. On this basis, a two-stage PDN restoration scheme is proposed that utilizes three emergency resources, including EVs, mobile energy storage systems (MESSs), and unmanned aerial vehicles (UAVs), to restore the power supply and communication of PDNs. This scheme first improves the distribution automation function, EV deployment capability, and traffic operation efficiency by prioritizing the recovery of communication network (CN) and urban traffic network (UTN) loads. Then, EVs and MESSs are further scheduled to achieve a better PDN restoration effect with the support of the restored CNs and UTNs. Case studies on a PDN, CN, and UTN integrated test system are conducted to verify the effectiveness of the proposed scheme. The results show that the prioritized load recovery operation for CN and UTN facilities in this scheme greatly improves the PDN restoration effect.

eess.SY↗

arXiv2024

Strategic SDN-based Microgrid Formation for Managing Communication Failures in Distribution System Restoration

Grid modernization has increased the reliance of power networks on cyber networks within distribution systems (DSs), heightening their vulnerability to disasters. Communication network failures significantly impede DS load recovery by diminishing observation and control. Prior research has largely ignored the need for integrated recovery of DS power and cyber networks' centralized control. Indeed, communication network restoration is critical for speedy load recovery through DS automation based microgrid formation. This paper exploits the data routing capabilities of software-defined networking (SDN) to enhance centralized control recovery in DS communication networks, incorporating it into a comprehensive DS restoration model. This model, tailored to the control requirements of load restoration, strategically allocates limited communication resources to re-establish connections between the operation center and terminal devices. Subsequently, DS automation is employed to orchestrate DS microgrid formation for power resupply. Additionally, we introduce a cyclic algorithm designed to optimize the load recovery via a multi-step, cooperative process. The efficacy of the proposed method is demonstrated on IEEE 33-node and IEEE 123-node test feeders.

eess.SY↗

arXiv2024

Resilient Microgrid Formation Considering Communication Interruptions

Distribution system (DS) communication failures following extreme events often degrade monitoring and control functions, thus preventing the acquisition of complete global DS component state information, on which existing post-disaster DS restoration methods are based. This letter proposes methods of inferring the states of DS components in the case of incomplete component state information. By using the known DS information, the operating states of unobservable DS branches and buses can be inferred, providing complete information for DS performance restoration before full communication recovery

eess.SY↗

arXiv2024

Secure and Scalable Network Slicing with Plug-and-Play Support for Power Distribution System Communication Networks

With the rapid development of power distribution systems (PDSs), the number of terminal devices and the types of delivered services involved are constantly growing. These trends make the operations of PDSs highly dependent on the support of advanced communication networks, which face two related challenges. The first is to provide sufficient flexibility, resilience, and security to meet varying demands and ensure the proper operation of gradually diversifying network services. The second is to realize the automatic identification of terminal devices, thus reducing the network maintenance burden. To solve these problems, this paper presents a novel multiservice network integration and device authentication slice-based network slicing scheme. In this scheme, the integration of PDS communication networks enables network resource sharing, and recovery from communication interruption is achieved through network slicing in the integrated network. Authentication servers periodically poll terminal devices, adjusting network slice ranges based on authentication results, thereby facilitating dynamic network slicing. Additionally, secure plug-and-play support for PDS terminal devices and network protection are achieved through device identification and dynamic adjustment of network slices. On this basis, a network optimization and upgrading methodology for load balancing and robustness enhancement is further proposed. This approach is designed to improve the performance of PDS communication networks, adapting to ongoing PDS development and the evolution of PDS services. The simulation results show that the proposed schemes endow a PDS communication network with favorable resource utilization, fault recovery, terminal device plug-and-play support, load balancing, and improved network robustness.

eess.SY↗

arXiv2024

Pay Attention to the Atlas: Atlas-Guided Test-Time Adaptation Method for Robust 3D Medical Image Segmentation

Convolutional neural networks (CNNs) often suffer from poor performance when tested on target data that differs from the training (source) data distribution, particularly in medical imaging applications where variations in imaging protocols across different clinical sites and scanners lead to different imaging appearances. However, re-accessing source training data for unsupervised domain adaptation or labeling additional test data for model fine-tuning can be difficult due to privacy issues and high labeling costs, respectively. To solve this problem, we propose a novel atlas-guided test-time adaptation (TTA) method for robust 3D medical image segmentation, called AdaAtlas. AdaAtlas only takes one single unlabeled test sample as input and adapts the segmentation network by minimizing an atlas-based loss. Specifically, the network is adapted so that its prediction after registration is aligned with the learned atlas in the atlas space, which helps to reduce anatomical segmentation errors at test time. In addition, different from most existing TTA methods which restrict the adaptation to batch normalization blocks in the segmentation network only, we further exploit the use of channel and spatial attention blocks for improved adaptability at test time. Extensive experiments on multiple datasets from different sites show that AdaAtlas with attention blocks adapted (AdaAtlas-Attention) achieves superior performance improvements, greatly outperforming other competitive TTA methods.

cs.CV↗