arXiv subjects

Hao Yang

Publications and source records attributed to Hao Yang.

At least 199 records · Page 11Linked to original sources

arXiv2024

"I've Heard of You!": Generate Spoken Named Entity Recognition Data for Unseen Entities

Spoken named entity recognition (NER) aims to identify named entities from speech, playing an important role in speech processing. New named entities appear every day, however, annotating their Spoken NER data is costly. In this paper, we demonstrate that existing Spoken NER systems perform poorly when dealing with previously unseen named entities. To tackle this challenge, we propose a method for generating Spoken NER data based on a named entity dictionary (NED) to reduce costs. Specifically, we first use a large language model (LLM) to generate sentences from the sampled named entities and then use a text-to-speech (TTS) system to generate the speech. Furthermore, we introduce a noise metric to filter out noisy data. To evaluate our approach, we release a novel Spoken NER benchmark along with a corresponding NED containing 8,853 entities. Experiment results show that our method achieves state-of-the-art (SOTA) performance in the in-domain, zero-shot domain adaptation, and fully zero-shot settings. Our data will be available at https://github.com/DeepLearnXMU/HeardU.

cs.CL↗

arXiv2024

M-Ped: Multi-Prompt Ensemble Decoding for Large Language Models

With the widespread application of Large Language Models (LLMs) in the field of Natural Language Processing (NLP), enhancing their performance has become a research hotspot. This paper presents a novel multi-prompt ensemble decoding approach designed to bolster the generation quality of LLMs by leveraging the aggregation of outcomes from multiple prompts. Given a unique input $X$, we submit $n$ variations of prompts with $X$ to LLMs in batch mode to decode and derive probability distributions. For each token prediction, we calculate the ensemble probability by averaging the $n$ probability distributions within the batch, utilizing this aggregated probability to generate the token. This technique is dubbed Inner-Batch Ensemble. To facilitate efficient batch inference, we implement a Left-Padding strategy to maintain uniform input lengths across the n prompts. Through extensive experimentation on diverse NLP tasks, including machine translation, code generation, and text simplification, we demonstrate the efficacy of our method in enhancing LLM performance. The results show substantial improvements in BLEU scores, pass@$k$ rates, and LENS metrics over conventional methods.

cs.CL↗

arXiv2024

Chain-of-Thought in Large Language Models: Decoding, Projection, and Activation

Chain-of-Thought prompting has significantly enhanced the reasoning capabilities of large language models, with numerous studies exploring factors influencing its performance. However, the underlying mechanisms remain poorly understood. To further demystify the operational principles, this work examines three key aspects: decoding, projection, and activation, aiming to elucidate the changes that occur within models when employing Chainof-Thought. Our findings reveal that LLMs effectively imitate exemplar formats while integrating them with their understanding of the question, exhibiting fluctuations in token logits during generation but ultimately producing a more concentrated logits distribution, and activating a broader set of neurons in the final layers, indicating more extensive knowledge retrieval compared to standard prompts. Our code and data will be publicly avialable when the paper is accepted.

cs.AI↗

arXiv2024

Data Uncertainty-Aware Learning for Multimodal Aspect-based Sentiment Analysis

As a fine-grained task, multimodal aspect-based sentiment analysis (MABSA) mainly focuses on identifying aspect-level sentiment information in the text-image pair. However, we observe that it is difficult to recognize the sentiment of aspects in low-quality samples, such as those with low-resolution images that tend to contain noise. And in the real world, the quality of data usually varies for different samples, such noise is called data uncertainty. But previous works for the MABSA task treat different quality samples with the same importance and ignored the influence of data uncertainty. In this paper, we propose a novel data uncertainty-aware multimodal aspect-based sentiment analysis approach, UA-MABSA, which weighted the loss of different samples by the data quality and difficulty. UA-MABSA adopts a novel quality assessment strategy that takes into account both the image quality and the aspect-based cross-modal relevance, thus enabling the model to pay more attention to high-quality and challenging samples. Extensive experiments show that our method achieves state-of-the-art (SOTA) performance on the Twitter-2015 dataset. Further analysis demonstrates the effectiveness of the quality assessment strategy.

cs.CL↗

arXiv2024

SCIGS: 3D Gaussians Splatting from a Snapshot Compressive Image

Snapshot Compressive Imaging (SCI) offers a possibility for capturing information in high-speed dynamic scenes, requiring efficient reconstruction method to recover scene information. Despite promising results, current deep learning-based and NeRF-based reconstruction methods face challenges: 1) deep learning-based reconstruction methods struggle to maintain 3D structural consistency within scenes, and 2) NeRF-based reconstruction methods still face limitations in handling dynamic scenes. To address these challenges, we propose SCIGS, a variant of 3DGS, and develop a primitive-level transformation network that utilizes camera pose stamps and Gaussian primitive coordinates as embedding vectors. This approach resolves the necessity of camera pose in vanilla 3DGS and enhances multi-view 3D structural consistency in dynamic scenes by utilizing transformed primitives. Additionally, a high-frequency filter is introduced to eliminate the artifacts generated during the transformation. The proposed SCIGS is the first to reconstruct a 3D explicit scene from a single compressed image, extending its application to dynamic 3D scenes. Experiments on both static and dynamic scenes demonstrate that SCIGS not only enhances SCI decoding but also outperforms current state-of-the-art methods in reconstructing dynamic 3D scenes from a single compressed image. The code will be made available upon publication.

cs.CV↗

arXiv2024

Optimized Vessel Segmentation: A Structure-Agnostic Approach with Small Vessel Enhancement and Morphological Correction

Accurate segmentation of blood vessels is essential for various clinical assessments and postoperative analyses. However, the inherent challenges of vascular imaging, such as sparsity, fine granularity, low contrast, data distribution variability, and the critical need for preserving topological structure, making generalized vessel segmentation particularly complex. While specialized segmentation methods have been developed for specific anatomical regions, their over-reliance on tailored models hinders broader applicability and generalization. General-purpose segmentation models introduced in medical imaging often fail to address critical vascular characteristics, including the connectivity of segmentation results. To overcome these limitations, we propose an optimized vessel segmentation framework: a structure-agnostic approach incorporating small vessel enhancement and morphological correction for multi-modality vessel segmentation. To train and validate this framework, we compiled a comprehensive multi-modality dataset spanning 17 datasets and benchmarked our model against six SAM-based methods and 17 expert models. The results demonstrate that our approach achieves superior segmentation accuracy, generalization, and a 34.6% improvement in connectivity, underscoring its clinical potential. An ablation study further validates the effectiveness of the proposed improvements. We will release the code and dataset at github following the publication of this work.

eess.IV↗

arXiv2024

Hard-Synth: Synthesizing Diverse Hard Samples for ASR using Zero-Shot TTS and LLM

Text-to-speech (TTS) models have been widely adopted to enhance automatic speech recognition (ASR) systems using text-only corpora, thereby reducing the cost of labeling real speech data. Existing research primarily utilizes additional text data and predefined speech styles supported by TTS models. In this paper, we propose Hard-Synth, a novel ASR data augmentation method that leverages large language models (LLMs) and advanced zero-shot TTS. Our approach employs LLMs to generate diverse in-domain text through rewriting, without relying on additional text data. Rather than using predefined speech styles, we introduce a hard prompt selection method with zero-shot TTS to clone speech styles that the ASR model finds challenging to recognize. Experiments demonstrate that Hard-Synth significantly enhances the Conformer model, achieving relative word error rate (WER) reductions of 6.5\%/4.4\% on LibriSpeech dev/test-other subsets. Additionally, we show that Hard-Synth is data-efficient and capable of reducing bias in ASR.

cs.CL↗

arXiv2024

Clustering and Ranking: Diversity-preserved Instruction Selection through Expert-aligned Quality Estimation

With contributions from the open-source community, a vast amount of instruction tuning (IT) data has emerged. Given the significant resource allocation required for training and evaluating models, it is advantageous to have an efficient method for selecting high-quality IT data. However, existing methods for instruction data selection have limitations such as relying on fragile external APIs, being affected by biases in GPT models, or reducing the diversity of the selected instruction dataset. In this paper, we propose an industrial-friendly, expert-aligned and diversity-preserved instruction data selection method: Clustering and Ranking (CaR). CaR employs a two-step process: first, it ranks instruction pairs using a high-accuracy (84.25%) scoring model aligned with expert preferences; second, it preserves dataset diversity through clustering. In our experiment, CaR efficiently selected a mere 1.96% of Alpaca's IT data, yet the resulting AlpaCaR model surpassed Alpaca's performance by an average of 32.1% in GPT-4 evaluations. Moreover, we find that data selecting is a consistent paradigm whether the pre-trained model is more capable or the model parameters scaling up. Our approach employs compact models with 550M parameters and incurs just 11.2% of the financial outlay of current methods, enhancing its industrial deployability.

cs.CL↗

arXiv2024

SynStitch: a Self-Supervised Learning Network for Ultrasound Image Stitching Using Synthetic Training Pairs and Indirect Supervision

Ultrasound (US) image stitching can expand the field-of-view (FOV) by combining multiple US images from varied probe positions. However, registering US images with only partially overlapping anatomical contents is a challenging task. In this work, we introduce SynStitch, a self-supervised framework designed for 2DUS stitching. SynStitch consists of a synthetic stitching pair generation module (SSPGM) and an image stitching module (ISM). SSPGM utilizes a patch-conditioned ControlNet to generate realistic 2DUS stitching pairs with known affine matrix from a single input image. ISM then utilizes this synthetic paired data to learn 2DUS stitching in a supervised manner. Our framework was evaluated against multiple leading methods on a kidney ultrasound dataset, demonstrating superior 2DUS stitching performance through both qualitative and quantitative analyses. The code will be made public upon acceptance of the paper.

eess.IV↗

arXiv2024

Series-to-Series Diffusion Bridge Model

Diffusion models have risen to prominence in time series forecasting, showcasing their robust capability to model complex data distributions. However, their effectiveness in deterministic predictions is often constrained by instability arising from their inherent stochasticity. In this paper, we revisit time series diffusion models and present a comprehensive framework that encompasses most existing diffusion-based methods. Building on this theoretical foundation, we propose a novel diffusion-based time series forecasting model, the Series-to-Series Diffusion Bridge Model ($\mathrm{S^2DBM}$), which leverages the Brownian Bridge process to reduce randomness in reverse estimations and improves accuracy by incorporating informative priors and conditions derived from historical time series data. Experimental results demonstrate that $\mathrm{S^2DBM}$ delivers superior performance in point-to-point forecasting and competes effectively with other diffusion-based models in probabilistic forecasting.

cs.LG↗

arXiv2024

Audio Is the Achilles' Heel: Red Teaming Audio Large Multimodal Models

Large Multimodal Models (LMMs) have demonstrated the ability to interact with humans under real-world conditions by combining Large Language Models (LLMs) and modality encoders to align multimodal information (visual and auditory) with text. However, such models raise new safety challenges of whether models that are safety-aligned on text also exhibit consistent safeguards for multimodal inputs. Despite recent safety-alignment research on vision LMMs, the safety of audio LMMs remains under-explored. In this work, we comprehensively red team the safety of five advanced audio LMMs under three settings: (i) harmful questions in both audio and text formats, (ii) harmful questions in text format accompanied by distracting non-speech audio, and (iii) speech-specific jailbreaks. Our results under these settings demonstrate that open-source audio LMMs suffer an average attack success rate of 69.14% on harmful audio questions, and exhibit safety vulnerabilities when distracted with non-speech audio noise. Our speech-specific jailbreaks on Gemini-1.5-Pro achieve an attack success rate of 70.67% on the harmful query benchmark. We provide insights on what could cause these reported safety-misalignments. Warning: this paper contains offensive examples.

cs.CL↗

arXiv2024

Double Mixture: Towards Continual Event Detection from Speech

Speech event detection is crucial for multimedia retrieval, involving the tagging of both semantic and acoustic events. Traditional ASR systems often overlook the interplay between these events, focusing solely on content, even though the interpretation of dialogue can vary with environmental context. This paper tackles two primary challenges in speech event detection: the continual integration of new events without forgetting previous ones, and the disentanglement of semantic from acoustic events. We introduce a new task, continual event detection from speech, for which we also provide two benchmark datasets. To address the challenges of catastrophic forgetting and effective disentanglement, we propose a novel method, 'Double Mixture.' This method merges speech expertise with robust memory mechanisms to enhance adaptability and prevent forgetting. Our comprehensive experiments show that this task presents significant challenges that are not effectively addressed by current state-of-the-art methods in either computer vision or natural language processing. Our approach achieves the lowest rates of forgetting and the highest levels of generalization, proving robust across various continual learning sequences. Our code and data are available at https://anonymous.4open.science/status/Continual-SpeechED-6461.

cs.CL↗

arXiv2024

From Handcrafted Features to LLMs: A Brief Survey for Machine Translation Quality Estimation

Machine Translation Quality Estimation (MTQE) is the task of estimating the quality of machine-translated text in real time without the need for reference translations, which is of great importance for the development of MT. After two decades of evolution, QE has yielded a wealth of results. This article provides a comprehensive overview of QE datasets, annotation methods, shared tasks, methodologies, challenges, and future research directions. It begins with an introduction to the background and significance of QE, followed by an explanation of the concepts and evaluation metrics for word-level QE, sentence-level QE, document-level QE, and explainable QE. The paper categorizes the methods developed throughout the history of QE into those based on handcrafted features, deep learning, and Large Language Models (LLMs), with a further division of deep learning-based methods into classic deep learning and those incorporating pre-trained language models (LMs). Additionally, the article details the advantages and limitations of each method and offers a straightforward comparison of different approaches. Finally, the paper discusses the current challenges in QE research and provides an outlook on future research directions.

cs.CL↗

arXiv2024

Tunable topological edge states in black phosphorus-like Bi(110)

We have investigated the structures and electronic properties of ultra-thin Bi(110) films grown on an s-wave superconductor substrate using low-temperature scanning tunneling microscopy and spectroscopy. Remarkably, our experimental results validate the theoretical predictions that the manipulation of Bi(110) surface atom buckling can control the topological phase transition. Notably, we have observed robust unreconstructed edge states at the edges of both 3-bilayer (BL) and 4-BL Bi(110) films, with the 4-BL film displaying stronger edge state intensity and a smaller degree of atomic buckling. First-principle calculations further substantiate these findings, demonstrating a gradual reduction in buckling as the film thickness increases, with average height differences between two Bi atoms of approximately 0.19 Å, 0.10 Å, 0.05 Å, and 0.00 Å for the 1-BL, 2-BL, 3-BL, and 4-BL Bi(110) films, respectively. When Bi films are larger than 2 layers, the system changes from a trivial to a non-trivial phase. This research sets the stage for the controlled realization of topological superconductors through the superconducting proximity effect, providing a significant platform for investigating Majorana zero modes and fabricating quantum devices.

cond-mat.mtrl-sci↗

arXiv2024

LMHaze: Intensity-aware Image Dehazing with a Large-scale Multi-intensity Real Haze Dataset

Image dehazing has drawn a significant attention in recent years. Learning-based methods usually require paired hazy and corresponding ground truth (haze-free) images for training. However, it is difficult to collect real-world image pairs, which prevents developments of existing methods. Although several works partially alleviate this issue by using synthetic datasets or small-scale real datasets. The haze intensity distribution bias and scene homogeneity in existing datasets limit the generalization ability of these methods, particularly when encountering images with previously unseen haze intensities. In this work, we present LMHaze, a large-scale, high-quality real-world dataset. LMHaze comprises paired hazy and haze-free images captured in diverse indoor and outdoor environments, spanning multiple scenarios and haze intensities. It contains over 5K high-resolution image pairs, surpassing the size of the biggest existing real-world dehazing dataset by over 25 times. Meanwhile, to better handle images with different haze intensities, we propose a mixture-of-experts model based on Mamba (MoE-Mamba) for dehazing, which dynamically adjusts the model parameters according to the haze intensity. Moreover, with our proposed dataset, we conduct a new large multimodal model (LMM)-based benchmark study to simulate human perception for evaluating dehazed images. Experiments demonstrate that LMHaze dataset improves the dehazing performance in real scenarios and our dehazing method provides better results compared to state-of-the-art methods.

cs.CV↗

arXiv2024

Jigsaw Puzzles: Splitting Harmful Questions to Jailbreak Large Language Models

Large language models (LLMs) have exhibited outstanding performance in engaging with humans and addressing complex questions by leveraging their vast implicit knowledge and robust reasoning capabilities. However, such models are vulnerable to jailbreak attacks, leading to the generation of harmful responses. Despite recent research on single-turn jailbreak strategies to facilitate the development of defence mechanisms, the challenge of revealing vulnerabilities under multi-turn setting remains relatively under-explored. In this work, we propose Jigsaw Puzzles (JSP), a straightforward yet effective multi-turn jailbreak strategy against the advanced LLMs. JSP splits questions into harmless fractions as the input of each turn, and requests LLMs to reconstruct and respond to questions under multi-turn interaction. Our experimental results demonstrate that the proposed JSP jailbreak bypasses original safeguards against explicitly harmful content, achieving an average attack success rate of 93.76% on 189 harmful queries across 5 advanced LLMs (Gemini-1.5-Pro, Llama-3.1-70B, GPT-4, GPT-4o, GPT-4o-mini). Moreover, JSP achieves a state-of-the-art attack success rate of 92% on GPT-4 on the harmful query benchmark, and exhibits strong resistant to defence strategies. Warning: this paper contains offensive examples.

cs.CL↗

arXiv2024

Semileptonic and nonleptonic weak decays of $ψ(1S,2S)$ and $η_{c}(1S,2S)$ to $D_{(s)}$ in the covariant light-front approach

In addition to the strong and electromagnetic decay modes, the $ψ(1S,2S)$ and $η_{c}(1S,2S)$ can also decay via the weak interaction. Such weak decays can be detected by the high-luminosity heavy-flavor experiments. At present, some of the semileptonic and nonleptonic $J/Ψ$ weak decays have been measured at BESIII. Researching for these charmonium weak decays to $D_{(s)}$ meson can provide a platform to check of the standard model (SM) and probe new physics (NP). So we investigate the semileptonic and nonleptonic weak decays of $ψ(1S,2S)$ and $η_{c}(1S,2S)$ to $D_{(s)}$ within the covariant light-front quark model (CLFQM). With form factors of the transitions $ψ(1S,2S)\to D_{(s)}$ and $η_{c}(1S,2S)\to D_{(s)}$ calculated under the CLFQM, we predict and discuss some physical observables, such as the branching ratios, the longitudinal polarizations $f_{L}$ and the forward-backward asymmetries $A_{FB}$. One can find that the Cabibbo-favored semi-leptonic decay channels $ψ(1S,2S)\to D_{s}^{-}\ell^{+}ν_{\ell}$ with $\ell=e,μ$ and the nonleptonic decay modes $ψ(1S,2S)\to D_{s}^{-}ρ^{+}$ have relatively large branching ratios of the order $\mathcal{O}(10^{-9})$, which are most likely to be accessible at the future high-luminosity experiments.

hep-ph↗

arXiv2024

Semileptonic and nonleptonic decays of $B_{u,d,s,c}^{*}$ in the covariant light-front approach

The semileptonic and nonleptonic decays of the b-flavor vector mesons $B^{*}_{u,d,s}$ and $B_{c}^{*}$ are investigated within the covariant light-front quark model (CLFQM). By calculating the form factors of the transitions $B_{u, d, s, c}^{*}\to P$ under the CLFQM, with $P$ denoting a pseudoscalar meson, i.e., $π, K, η_c(1S,2S), D_{(s)}, B_{(s)}$, we predict and discuss several physical observables, including the branching ratios, polarization fractions $f_{L}, f_{\|}$, and forward-backward asymmetries $A_{FB}$. The total widths of the single-photon radiative decay channels for these b-flavor vector mesons are estimated using their partial widths. In these considered decays, one can find that the semileptonic decays $B_{s}^{*0}\to D_{s}^{-}\ell^{\prime+}ν_{\ell^\prime}$ and $B_{c}^{*+}\to B_{s}^{0}\ell^{\prime+}ν_{\ell^\prime}, η_{c}\ell^{\prime+}ν_{\ell^\prime}$, with $\ell^\prime$ being $e$ or $τ$, and the nonleptonic channels $B_{c}^{*+}\to B^0_{s} π^{+}, B^0_{s} ρ^{+}$ have the largest branching ratios, which can reach up to the $10^{-7}$ order, and are most likely to be accessible at the future high-luminosity LHCb and Belle-II experiments.

hep-ph↗