Search

This paper presents the submission of Huawei Translation Services Center (HW-TSC) to machine translation tasks of the 20th China Conference on Machine Translation (CCMT 2024). We participate in the bilingual machine translation task and multi-domain machine translation task. For these two translation tasks, we use training strategies such as regularized dropout, bidirectional training, data diversification, forward translation, back translation, alternated training, curriculum learning, and transductive ensemble learning to train neural machine translation (NMT) models based on the deep Transformer-big architecture. Furthermore, to explore whether large language model (LLM) can help improve the translation quality of NMT systems, we use supervised fine-tuning to train llama2-13b as an Automatic post-editing (APE) model to improve the translation results of the NMT model on the multi-domain machine translation task. By using these plyometric strategies, our submission achieves a competitive result in the final evaluation., Comment: 14 pages, 2 figures, 6 Tables, CCMT2024. arXiv admin note: substantial text overlap with arXiv:2409.14800

This paper presents the submission of Huawei Translate Services Center (HW-TSC) to the WMT24 general machine translation (MT) shared task, where we participate in the English to Chinese (en2zh) language pair. Similar to previous years' work, we use training strategies such as regularized dropout, bidirectional training, data diversification, forward translation, back translation, alternated training, curriculum learning, and transductive ensemble learning to train the neural machine translation (NMT) model based on the deep Transformer-big architecture. The difference is that we also use continue pre-training, supervised fine-tuning, and contrastive preference optimization to train the large language model (LLM) based MT model. By using Minimum Bayesian risk (MBR) decoding to select the final translation from multiple hypotheses for NMT and LLM-based MT models, our submission receives competitive results in the final evaluation., Comment: 10 pages, 4 figures, 2 Tables, EMNLP2024

Large pre-trained models (LPMs) have demonstrated exceptional performance in diverse natural language processing and computer vision tasks. However, fully fine-tuning these models poses substantial memory challenges, particularly in resource-constrained environments. Parameter-efficient fine-tuning (PEFT) methods, such as LoRA, mitigate this issue by adjusting only a small subset of parameters. Nevertheless, these methods typically employ random initialization for low-rank matrices, which can lead to inefficiencies in gradient descent and diminished generalizability due to suboptimal starting points. To address these limitations, we propose SVFit, a novel PEFT approach that leverages singular value decomposition (SVD) to initialize low-rank matrices using critical singular values as trainable parameters. Specifically, SVFit performs SVD on the pre-trained weight matrix to obtain the best rank-r approximation matrix, emphasizing the most critical singular values that capture over 99% of the matrix's information. These top-r singular values are then used as trainable parameters to scale the fundamental subspaces of the matrix, facilitating rapid domain adaptation. Extensive experiments across various pre-trained models in natural language understanding, text-to-image generation, and image classification tasks reveal that SVFit outperforms LoRA while requiring 16 times fewer trainable parameters.

A timeline provides a total ordering of events and times, and is useful for a number of natural language understanding tasks. However, qualitative temporal graphs that can be derived directly from text -- such as TimeML annotations -- usually explicitly reveal only partial orderings of events and times. In this work, we apply prior work on solving point algebra problems to the task of extracting timelines from TimeML annotated texts, and develop an exact, end-to-end solution which we call TLEX (TimeLine EXtraction). TLEX transforms TimeML annotations into a collection of timelines arranged in a trunk-and-branch structure. Like what has been done in prior work, TLEX checks the consistency of the temporal graph and solves it; however, it adds two novel functionalities. First, it identifies specific relations involved in an inconsistency (which could then be manually corrected) and, second, TLEX performs a novel identification of sections of the timelines that have indeterminate order, information critical for downstream tasks such as aligning events from different timelines. We provide detailed descriptions and analysis of the algorithmic components in TLEX, and conduct experimental evaluations by applying TLEX to 385 TimeML annotated texts from four corpora. We show that 123 of the texts are inconsistent, 181 of them have more than one ``real world'' or main timeline, and there are 2,541 indeterminate sections across all four corpora. A sampling evaluation showed that TLEX is 98--100% accurate with 95% confidence along five dimensions: the ordering of time-points, the number of main timelines, the placement of time-points on main versus subordinate timelines, the connecting point of branch timelines, and the location of the indeterminate sections. We provide a reference implementation of TLEX, the extracted timelines for all texts, and the manual corrections of the inconsistent texts., Comment: 25 pages, 9 figures

Despite the subjective nature of semantic textual similarity (STS) and pervasive disagreements in STS annotation, existing benchmarks have used averaged human ratings as the gold standard. Averaging masks the true distribution of human opinions on examples of low agreement, and prevents models from capturing the semantic vagueness that the individual ratings represent. In this work, we introduce USTS, the first Uncertainty-aware STS dataset with ~15,000 Chinese sentence pairs and 150,000 labels, to study collective human opinions in STS. Analysis reveals that neither a scalar nor a single Gaussian fits a set of observed judgements adequately. We further show that current STS models cannot capture the variance caused by human disagreement on individual instances, but rather reflect the predictive confidence over the aggregate dataset., Comment: 16 pages, 7 figures

Recently, the Hellings Downs correlation has been observed by different pulsar timing array (PTA) collaborations, such as NANOGrav, European PTA, Parkes PTA, and Chinese PTA. These PTA measurements of the most precise pulsars within the Milky Way show the first evidence for the stochastic gravitational wave background of our Universe. We study the ultralight axion interpretation of the new discovery by investigating the gravitational wave from axion transitions between different energy levels of the gravitational atoms, which are composed of cosmic populated Kerr black holes and their surrounding axion clouds formed from the superradiant process. We demonstrate that this new observation naturally admits an ultralight axion interpretation around $2\times10^{-21}$~eV, which is consistent with the typical fuzzy dark matter., Comment: The Bayesian analysis and contributions of supermassive blackhole binaries added, transition energy level corrected

Virtual reality (VR) produces a highly realistic simulated environment with controllable environment variables. This paper proposes a Dynamic Scene Adjustment (DSA) mechanism based on the user interaction status and performance, which aims to adjust the VR experiment variables to improve the user's game engagement. We combined the DSA mechanism with a musical rhythm VR game. The experimental results show that the DSA mechanism can improve the user's game engagement (task performance).

3D cross-modal retrieval is gaining attention in the multimedia community. Central to this topic is learning a joint embedding space to represent data from different modalities, such as images, 3D point clouds, and polygon meshes, to extract modality-invariant and discriminative features. Hence, the performance of cross-modal retrieval methods heavily depends on the representational capacity of this embedding space. Existing methods treat all instances equally, applying the same penalty strength to instances with varying degrees of difficulty, ignoring the differences between instances. This can result in ambiguous convergence or local optima, severely compromising the separability of the feature space. To address this limitation, we propose an Instance-Variant loss to assign different penalty strengths to different instances, improving the space separability. Specifically, we assign different penalty weights to instances positively related to their intra-class distance. Simultaneously, we reduce the cross-modal discrepancy between features by learning a shared weight vector for the same class data from different modalities. By leveraging the Gaussian RBF kernel to evaluate sample similarity, we further propose an Intra-Class loss function that minimizes the intra-class distance among same-class instances. Extensive experiments on three 3D cross-modal datasets show that our proposed method surpasses recent state-of-the-art approaches.

The rapid expansion of the Internet of Things (IoT) and Edge Computing has presented challenges for centralized Machine and Deep Learning (ML/DL) methods due to the presence of distributed data silos that hold sensitive information. To address concerns regarding data privacy, collaborative and privacy-preserving ML/DL techniques like Federated Learning (FL) have emerged. However, ensuring data privacy and performance alone is insufficient since there is a growing need to establish trust in model predictions. Existing literature has proposed various approaches on trustworthy ML/DL (excluding data privacy), identifying robustness, fairness, explainability, and accountability as important pillars. Nevertheless, further research is required to identify trustworthiness pillars and evaluation metrics specifically relevant to FL models, as well as to develop solutions that can compute the trustworthiness level of FL models. This work examines the existing requirements for evaluating trustworthiness in FL and introduces a comprehensive taxonomy consisting of six pillars (privacy, robustness, fairness, explainability, accountability, and federation), along with over 30 metrics for computing the trustworthiness of FL models. Subsequently, an algorithm named FederatedTrust is designed based on the pillars and metrics identified in the taxonomy to compute the trustworthiness score of FL models. A prototype of FederatedTrust is implemented and integrated into the learning process of FederatedScope, a well-established FL framework. Finally, five experiments are conducted using different configurations of FederatedScope to demonstrate the utility of FederatedTrust in computing the trustworthiness of FL models. Three experiments employ the FEMNIST dataset, and two utilize the N-BaIoT dataset considering a real-world IoT security use case.

Current and imminent quantum hardware lacks reliability and applicability due to noise and limited qubit counts. Quantum circuit cutting -- a technique dividing large quantum circuits into smaller subcircuits with sizes appropriate for the limited quantum resource at hand -- is used to mitigate these problems. However, classical postprocessing involved in circuit cutting generally grows exponentially with the number of cuts and quantum counts. This article introduces the notion of approximate circuit reconstruction. Using a sampling-based method like Markov Chain Monte Carlo (MCMC), we probabilistically select bit strings of high probability upon reconstruction. This avoids excessive calculations when reconstructing the full probability distribution. Our results show that such a sampling-based postprocessing method holds great potential for fast and reliable circuit reconstruction in the NISQ era and beyond.

The detection of gravitational wave (GW) has opened a new window to test the theory of gravity in the strong field regime. In general relativity (GR), GW can only possess two tensor polarization modes, which are known as the $+$ and $\times$ modes. However, vector and scalar modes can exist in some modified theories of gravity, and we can test the gravitational theories by probing these extra polarization modes. As a space-borne GW detector, TianQin will be launched in the 2030s, and it is expected to observe plenty of GW signals, including those from nearly ten-thousands double white dwarfs (DWDs) in our galaxy. This offers an excellent chance for testing the existence of extra polarization modes. In this article, we analyze the capability of detecting the extra polarization modes with DWDs. For the extra modes, we consider both the leading order dipole radiation and the subleading quadrupole radiation. We find that the capability of TianQin has very strong dependency on the source orientation. We also analyze all the verification binaries with determined position and frequency, and find that ZTF J1539 can give the best constraint on the extra polarization modes. We have also considered the case of TianQin twin constellation, and the joint observation with LISA., Comment: 13 pages, 18 figures, 5 tables, matches the published version

The axion or axion-like particle motivated from a natural solution of strong CP problem or string theory is a promising dark matter candidate. We study the new observational effects of ultralight axion-like particles by the space-borne gravitational wave detector and the radio telescope. Taking the neutron star-black hole binary as an example, we demonstrate that the gravitational waveform could be obviously modified by the slow depletion of the axion cloud around the black hole formed through the superradiance process. We compare these new effects on the binary with the well-studied effects from dynamical friction with dark matter and dipole radiation in model-independent ways. Finally, we discuss the constraints from LIGO/Virgo and study the detectability of the ultralight axion particles at LISA and TianQin., Comment: Published version at SCIENCE CHINA Physics, Mechanics & Astronomy

Determinantal Point Processes (DPPs) are a widely used probabilistic model for negatively correlated sets. DPPs have been successfully employed in Machine Learning applications to select a diverse, yet representative subset of data. In seminal work on DPPs in Machine Learning, Kulesza conjectured in his PhD Thesis (2011) that the problem of finding a maximum likelihood DPP model for a given data set is NP-complete. In this work we prove Kulesza's conjecture. In fact, we prove the following stronger hardness of approximation result: even computing a $\left(1-O(\frac{1}{\log^9{N}})\right)$-approximation to the maximum log-likelihood of a DPP on a ground set of $N$ elements is NP-complete. At the same time, we also obtain the first polynomial-time algorithm that achieves a nontrivial worst-case approximation to the optimal log-likelihood: the approximation factor is $\frac{1}{(1+o(1))\log{m}}$ unconditionally (for data sets that consist of $m$ subsets), and can be improved to $1-\frac{1+o(1)}{\log N}$ if all $N$ elements appear in a $O(1/N)$-fraction of the subsets. In terms of techniques, we reduce approximating the maximum log-likelihood of DPPs on a data set to solving a gap instance of a "vector coloring" problem on a hypergraph. Such a hypergraph is built on a bounded-degree graph construction of Bogdanov, Obata and Trevisan (FOCS 2002), and is further enhanced by the strong expanders of Alon and Capalbo (FOCS 2007) to serve our purposes.

Quantitative analysis of Game User eXperience (GUX) is important to the game industry. Different from the typical questionnaire analysis, this paper focuses on the computational analysis of GUX. We aim to analyze the relationship between game and players using the multi-modal data including physiological data and game process data. We theoretically extend the Flow model from the classic skill-and-challenge plane by expanding new dimension on motivation, which is the result of the multi-modal data analysis on affect, and physiological data. We call this 3D Flow as Motivational Flow, MovFlow. Meanwhile, we implement a quantitative GUX Analysis System (GUXAS), which can predict the player's in-game experience state by only using game process data. It analyzes the correlation among not only in-game state, but the player's psychological-and-physiological reaction in the entire interactive game-play process. The experiments demonstrated our MovFlow model efficiently distinguished the users' in-game experience states from the perspective of GUX.

To adapt to different environments, human circadian rhythms will be constantly adjusted as the environment changes, which follows the principle of survival of the fittest. According to this principle, objective factors (such as circadian rhythms, and light intensity) can be utilized to control time perception. The subjective judgment on the estimation of elapsed time is called time perception. In the physical world, factors that can affect time perception, represented by illumination, are called the Zeitgebers. In recent years, with the development of Virtual Reality (VR) technology, effective control of zeitgebers has become possible, which is difficult to achieve in the physical world. Based on previous studies, this paper deeply explores the actual performance in VR environment of four types of time zeitgebers (music, color, cognitive load, and concentration) that have been proven to have a certain impact on time perception in the physical world. It discusses the study of the measurement of the difference between human time perception and objective escaped time in the physical world.

Modern Web systems such as social media and e-commerce contain rich contents expressed in images and text. Leveraging information from multi-modalities can improve the performance of machine learning tasks such as classification and recommendation. In this paper, we propose the Cross-Modality Attention Contrastive Language-Image Pre-training (CMA-CLIP), a new framework which unifies two types of cross-modality attentions, sequence-wise attention and modality-wise attention, to effectively fuse information from image and text pairs. The sequence-wise attention enables the framework to capture the fine-grained relationship between image patches and text tokens, while the modality-wise attention weighs each modality by its relevance to the downstream tasks. In addition, by adding task specific modality-wise attentions and multilayer perceptrons, our proposed framework is capable of performing multi-task classification with multi-modalities. We conduct experiments on a Major Retail Website Product Attribute (MRWPA) dataset and two public datasets, Food101 and Fashion-Gen. The results show that CMA-CLIP outperforms the pre-trained and fine-tuned CLIP by an average of 11.9% in recall at the same level of precision on the MRWPA dataset for multi-task classification. It also surpasses the state-of-the-art method on Fashion-Gen Dataset by 5.5% in accuracy and achieves competitive performance on Food101 Dataset. Through detailed ablation studies, we further demonstrate the effectiveness of both cross-modality attention modules and our method's robustness against noise in image and text inputs, which is a common challenge in practice., Comment: 9 pages, 2 figures, 6 tables, 1 algorithm

To accommodate the advent of microgrids (MG) managing distributed energy resources (DER) in distribution systems, an interactive two-stage joint retail electricity market mechanism is proposed to provide an effective platform for these prosumers to proactively join in retail transactions. Day-ahead stochastic energy trading between the distribution system operator (DSO) and MGs is conducted in the first stage of a centralized retail market, where a chance-constrained uncertainty distribution locational marginal price (CC-UDLMP) containing the cost of uncertainty precautions is used to settle transactions. In the second stage, a novel intra-day peer-to-peer-based (P2P) flexibility transaction pattern is implemented between MGs in local flexibility markets under the regulation of DSO to eliminate power imbalances caused by rolling-based estimates whilst considering systematic operations. A fully distributed iterative algorithm is presented to find the equilibrium solution of this two-stage sequential game framework. Moreover, in order to enhance the versatility of this algorithm, an improved Lp-box alternating direction methods of multipliers (ADMM) algorithm is used to efficiently resolve the first-stage stochastic economic dispatch problem with a mixed-integer second-order cone structure. It is verified that the proposed market mechanism can effectively improve the overall market efficiency under uncertainties.