Performance Efficiency Trade-off Research Articles

CAM: A cross-lingual adaptation framework for low-resource language speech recognition

In this paper, a novel cross-lingual adaptation framework called CAM is presented for low-resource language speech recognition (LLSR). It is based on the recent popular adapter method. CAM is achieved by adapting self-supervised speech models (SSMs) from source languages to target low-resource languages in a two-stage process. CAM fills two research gaps existing in current methods: (i) language similarity is not effectively considered; (ii) the performance-efficiency trade-off is not well balanced. Specifically, two key components, a similarity-aware fusion module (SAFM) and an adapter weight-sharing strategy (AWSS), are designed. A well-trained adapter is introduced by SAFM to compute precise language similarities via dot product. AWSS takes both performance and efficiency into consideration by sharing adapter weights. Experimental results on two corpora, FLEURS and Common Voice, demonstrate that CAM equipped with these two designs, denoted as performance-oriented CAM (P-CAM), obtains state-of-the-art (SOTA) performance with satisfactory efficiency compared to current leading methods. Besides, an efficiency-oriented CAM (E-CAM) by introducing a weight-space fusion module (WSFM) is presented. The core of WSFM is to average the weights of multiple adapters into a new adapter. Compared to full fine-tuning, only 5.0% of trainable parameters are required for E-CAM while a 2.3% relative average word error rate (WER) reduction is obtained. CAM brings a performance- or efficiency-oriented trade-off to meet the different needs of speech recognition systems.

AutoPEFT: Automatic Configuration Search for Parameter-Efficient Fine-Tuning

<scp>AutoPEFT</scp>: Automatic Configuration Search for Parameter-Efficient Fine-Tuning

Super-efficient Echocardiography Video Segmentation via Proxy- and Kernel-Based Semi-supervised Learning

Automatic segmentation of left ventricular endocardium in echocardiography videos is critical for assessing various cardiac functions and improving the diagnosis of cardiac diseases. It is yet a challenging task due to heavy speckle noise, significant shape variability of cardiac structure, and limited labeled data. Particularly, the real-time demand in clinical practice makes this task even harder. In this paper, we propose a novel proxy- and kernel-based semi-supervised segmentation network (PKEcho-Net) to comprehensively address these challenges. We first propose a multi-scale region proxy (MRP) mechanism to model the region-wise contexts, in which a learnable region proxy with an arbitrary shape is developed in each layer of the encoder, allowing the network to identify homogeneous semantics and hence alleviate the influence of speckle noise on segmentation. To sufficiently and efficiently exploit temporal consistency, different from traditional methods which only utilize the temporal contexts of two neighboring frames via feature warping or self-attention mechanism, we formulate the semi-supervised segmentation with a group of learnable kernels, which can naturally and uniformly encode the appearances of left ventricular endocardium, as well as extracting the inter-frame contexts across the whole video to resist the fast shape variability of cardiac structures. Extensive experiments have been conducted on two famous public echocardiography video datasets, EchoNet-Dynamic and CAMUS. Our model achieves the best performance-efficiency trade-off when compared with other state-of-the-art approaches, attaining comparative accuracy with a much faster speed. The code is available at https://github.com/JingyinLin/PKEcho-Net.

Searching part-specific neural fabrics for human pose estimation

Neural architecture search (NAS) has emerged in many domains to jointly learn the architectures and weights of neural networks. The core spirit behind NAS is to automatically search neural architectures for target tasks with better performance-efficiency trade-offs. However, existing approaches emphasize on only searching a single architecture with less human intervention to replace a human-designed neural network, yet making the search process almost independent of the domain knowledge. In this paper, we aim to apply NAS for human pose estimation and we ask: when NAS meets this localization task, can the articulated human body structure help to search better task-specific architectures?To this end, we first design a new neural architecture search space, Cell-based Neural Fabric (CNF), to learn micro as well as macro neural architecture using a differentiable search strategy. Then, by viewing locating human parts as multiple disentangled prediction sub-tasks, we exploit the compositionality of human body structure as guidance to search multiple part-specific CNFs specialized for different human parts. After the search, all these part-specific neural fabrics have been tailored with distinct micro and macro architecture parameters. The results show that such knowledge-guided NAS-based model outperforms a hand-crafted part-based baseline model, and the resulting multiple part-specific architectures gain significant performance improvement against a single NAS-based architecture for the whole body. The experiments on MPII and COCO datasets show that our models11Code is available at https://github.com/yangsenius/PoseNFS. achieve comparable performance against the state-of-the-art methods while being relatively lightweight.

Network Structure, Efficiency, and Performance in WikiProjects

The internet has enabled collaborations at a scale never before possible, but the best practices for organizing such large collaborations are still not clear. Wikipedia is a visible and successful example of such a collaboration which might offer insight into what makes large-scale, decentralized collaborations successful. We analyze the relationship between the structural properties of WikiProject coeditor networks and the performance and efficiency of those projects. We confirm the existence of an overall performance-efficiency trade-off, while observing that some projects are higher than others in both performance and efficiency, suggesting the existence factors correlating positively with both. Namely, we find an association between low-degree coeditor networks and both high performance and high efficiency. We also confirm results seen in previous numerical and small-scale lab studies: higher performance with less skewed node distributions, and higher performance with shorter path lengths. We use agent-based models to explore possible mechanisms for degree-dependent performance and efficiency. We present a novel local-majority learning strategy designed to satisfy properties of real-world collaborations. The local-majority strategy as well as a localized conformity-based strategy both show degree-dependent performance and efficiency, but in opposite directions, suggesting that these factors depend on both network structure and learning strategy. Our results suggest possible benefits to decentralized collaborations made of smaller, more tightly-knit teams, and that these benefits may be modulated by the particular learning strategies in use.

Geographic Image Retrieval Using Local Invariant Features

This paper investigates local invariant features for geographic (overhead) image retrieval. Local features are particularly well suited for the newer generations of aerial and satellite imagery whose increased spatial resolution, often just tens of centimeters per pixel, allows a greater range of objects and spatial patterns to be recognized than ever before. Local invariant features have been successfully applied to a broad range of computer vision problems and, as such, are receiving increased attention from the remote sensing community particularly for challenging tasks such as detection and classification. We perform an extensive evaluation of local invariant features for image retrieval of land-use/land-cover (LULC) classes in high-resolution aerial imagery. We report on the effects of a number of design parameters on a bag-of-visual-words (BOVW) representation including saliency- versus grid-based local feature extraction, the size of the visual codebook, the clustering algorithm used to create the codebook, and the dissimilarity measure used to compare the BOVW representations. We also perform comparisons with standard features such as color and texture. The performance is quantitatively evaluated using a first-of-its-kind LULC ground truth data set which will be made publicly available to other researchers. In addition to reporting on the effects of the core design parameters, we also describe interesting findings such as the performance-efficiency tradeoffs that are possible through the appropriate pairings of different-sized codebooks and dissimilarity measures. While the focus is on image retrieval, we expect our insights to be informative for other applications such as detection and classification.