arXiv subjects

Xinyu Liang

Publications and source records attributed to Xinyu Liang.

At least 19 recordsLinked to original sources

arXiv2026

SimulS2ST-Omni: Data-Efficient Streaming Speech-to-Speech Translation via Explicit Trajectory Supervision

Long-form streaming speech-to-speech translation (S2ST) requires incremental, unbounded translation under strict latency constraints. Existing methods typically suffer from sentence-bounded supervision or demand massive paired-S2ST supervision. We introduce a training recipe enabling a speech language model for sentence-level and long-form streaming S2ST using only $\sim$2k hours of paired cross-lingual S2ST data, layered atop auxiliary supervision. Anchored by auxiliary multitask training, our approach remains robust even when the paired-S2ST budget itself is reduced by 90\%. Our core contribution, joint text-code trajectory supervision, schedules target text and acoustic semantic codes as a unified commitment path, eliminating the need for separate, unstable speech-side emission controllers. Furthermore, our two-stream Thinker--Talker factorization significantly outperforms unified-decoder baselines by decoupling linguistic reasoning from dense acoustic prediction to mitigate modality interference. Finally, our system achieves highly competitive quality-latency trade-offs on RealSI and ACL60/60-dev, matching state-of-the-art, closed-source S2ST systems such as LiveInterpret~2.0 on ASR-BLEU.

cs.SD↗

arXiv2026

CHILDES-Aligned: A Curated Children's Speech Dataset via Multi-Model Timestamp Ensembling

CHILDES is a large-scale child speech corpus containing long-form recordings of naturalistic child-adult interactions, making it a valuable resource for studying child speech and language development. However, utterance-level timestamps provided in this corpus are often noisy, incomplete, or misaligned with the audio. As a result, utterances cannot always be reliably localized within long recordings, which limits the direct use of these data for training and evaluating speech models. In this work, we propose BEACON (Boundary Estimation via Alignment CONsensus), an ensemble timestamp-curation framework that refines utterance-level timestamps by aggregating knowledge from multiple off-the-shelf ASR models. Specifically, each model's word-level timestamp predictions are first aligned to provided human transcripts, and the final utterance time boundaries are determined by a consensus voting strategy. The framework is corpus-agnostic and applies to any long-form recording paired with a trusted transcript whose timestamps are unreliable or missing, offering a general recipe for timestamp curation. Leveraging this pipeline, we curate and release a 413-hour general-purpose child-speech dataset with corrected utterance-level timestamps, together with a 283-hour quality-controlled subset for ASR training. Fine-tuning on this subset yields up to an average 19.5% relative WER reduction on four out-of-domain child-speech benchmarks.

eess.AS↗

arXiv2026

FlexiSLM: A Spoken Language Model with Dynamic and Controllable Frame Rates

Spoken language models (SLMs) extend LLMs to speech input and output. Existing SLMs represent speech at fixed frame rates (e.g., 25 or 12.5 Hz), ignoring the time-varying information density of speech and offering limited flexibility to trade off quality for speed at inference time. Recent audio tokenizer research has proposed dynamic-frame-rate speech coding, which exploits this non-uniformity and enables two new capabilities: very low average frame rates and frame-rate controllability. However, this technique has not yet been applied to SLMs. We introduce FlexiSLM, the first SLM with dynamic and controllable frame rates. FlexiSLM uses the pretrained FlexiCodec to obtain dynamic speech output tokens. The main contributions of this work are threefold: (1) integrating and validating this dynamic-rate representation within a multi-task, speech-to-speech SLM architecture; (2) extending it to frame compression on the input side; and (3) introducing direct frame-rate conditioning to enable accurate and controllable SLM inference. FlexiSLM outperforms fixed-frame-rate 7B models including Qwen2.5-Omni and Kimi-Audio at its 12.5 Hz and 6.25 Hz operating points. We further verify that FlexiSLM can be accurately steered down to 4.0 Hz; at 6.25 Hz, it roughly halves inference time relative to 12.5 Hz while retaining strong speech-to-speech quality. Audio samples are available at: https://flexislm.github.io. Code and data are available at: https://github.com/AmphionTeam/FlexiSLM.

cs.SD↗

arXiv2026

DNSMOS-C: Improving End-to-end Speech Quality Models via Contrastive Learning

We introduce DNSMOS-C, a compact end-to-end speech quality assessment model that extends the DNSMOS Pro framework by integrating a MOS-guided triplet-based contrastive loss. Applied directly to the intermediate embeddings, this contrastive supervision encourages the latent space to be better organized with respect to perceptual quality while preserving the simplicity and efficiency of DNSMOS Pro. Unlike prior methods that depend on large pre-trained self-supervised learning (SSL) encoders and multi-stage training, DNSMOS-C jointly learns speech representations and MOS regression within a single, unified framework. Experiments on multiple datasets show that DNSMOS-C consistently improves correlation metrics over DNSMOS Pro and achieves better generalization on challenging out-of-domain test sets. Furthermore, latent space analyses indicate that our approach learns representations that exhibit an emergent low-dimensional quality ordering, which enhances interpretability and improves training stability. These findings demonstrate that MOS-guided contrastive learning enables more robust and accurate quality predictions without incurring additional computational overhead.

eess.AS↗

arXiv2026

Direct observation of quadruple spin-texture locking in a 2D d-wave altermagnet

Altermagnets combine vanishing net magnetization with nonrelativistic, momentum-dependent spin splitting, offering a new paradigm for spintronics. Spin-crystal symmetry coupling, namely spin-lattice locking, is the defining mechanism of altermagnetism, enforcing opposite spin sublattices in real space and spin-momentum-locked electronic structure in reciprocal space. Direct atomic-scale visualization of spin-lattice locking therefore constitutes a decisive benchmark of the altermagnetic state, yet such evidence has remained elusive despite extensive efforts. Here we show that the electronic states in RbV2Se2O exhibit a d-wave-like spin texture at the sublattice level, providing the first atomic-scale evidence of spin-lattice locking with a predominantly c-axis spin orientation. By employing an in-situ, field-switchable spin-polarized Cr tip, we realize spin-contrast mapping of quasiparticle interference at identical energies, overcoming a long-standing experimental barrier in altermagnets. The resulting interference patterns exhibit pronounced spin-dependent modulations, establishing spin scattering locking and spin momentum locking as the real and reciprocal space manifestations. Unexpectedly, we uncover that the spin-selective scattering response is organized by a long-period stripe modulation, giving rise to a previously unidentified form of spin-texture locking, spin-stripe locking. We attribute this behavior to the emergence of a spin-density-wave moiré pattern. Together, these results establish a unified picture of quadruple spin-texture locking phenomena in a d-wave altermagnet, and position altermagnets as a versatile platform for exploring many-body interactions among intertwined degrees of freedom, including spin, lattice, momentum, moiré potential and valley.

cond-mat.mtrl-sci↗

arXiv2026

SA-SSL-MOS: Self-supervised Learning MOS Prediction with Spectral Augmentation for Generalized Multi-Rate Speech Assessment

Designing a speech quality assessment (SQA) system for estimating mean-opinion-score (MOS) of multi-rate speech with varying sampling frequency (16-48 kHz) is a challenging task. The challenge arises due to the limited availability of a MOS-labeled training dataset comprising multi-rate speech samples. While self-supervised learning (SSL) models have been widely adopted in SQA to boost performance, a key limitation is that they are pretrained on 16 kHz speech and therefore discard high-frequency information present in higher sampling rates. To address this issue, we propose a spectrogram-augmented SSL method that incorporates high-frequency features (up to 48 kHz sampling rate) through a parallel-branch architecture. We further introduce a two-step training scheme: the model is first pre-trained on a large 48 kHz dataset and then fine-tuned on a smaller multi-rate dataset. Experimental results show that leveraging high-frequency information overlooked by SSL features is crucial for accurate multi-rate SQA, and that the proposed two-step training substantially improves generalization when multi-rate data is limited.

eess.AS↗

arXiv2025

Novel coupling between charge order and time-reversal-symmetry-breaking superconductivity

The interplay between charge-density waves (CDWs), which break translational symmetry, and spatially homogeneous superconductivity, which breaks global U(1) gauge symmetry, can give rise to an intriguing phenomenon: the pair-density wave, characterized by a spatial modulation of the superconducting order parameter. Yet how CDWs couple to unconventional superconducting states-particularly those with time-reversal symmetry breaking (TRSB)-remains largely unexplored. Here, using scanning tunneling microscopy on heavily hole-doped Ba$_{1-x}$K$_x$Fe$_2$As$_2$, which hosts an s $\pm$ is superconducting state, we reveal a previously unobserved coupling between a surface CDW and TRSB superconductivity. Experimentally, the TRSB superconductivity imparts "chirality" to the CDW, which manifests as commensurate domains separated by domain walls with $π$-phase slips-forming what we term a bipolar CDW. The domain walls delineate TRSB domains of opposite chirality, consistent with spontaneous breaking of U(1) $\times$ Z2. Supported by theoretical modelling, we construct a framework in which a hidden interfacial pair-density modulation (PDM) mediates a linear coupling between the surface CDW and interband Josephson currents of TRSB superconductivity. Crucially, the theory shows that realizing this linear coupling requires a controlled global phase difference $δ$ $ϕ$ = $π$/2 between the PDM and CDW states. Our results uncover a previously overlooked connection between charge ordering and TRSB superconductivity, opening a pathway to explore intertwined quantum orders in iron-based superconductors and other strongly correlated systems.

cond-mat.supr-con↗

arXiv2025

Selection of Layers from Self-supervised Learning Models for Predicting Mean-Opinion-Score of Speech

Self-supervised learning (SSL) models like Wav2Vec2, HuBERT, and WavLM have been widely used in speech processing. These transformer-based models consist of multiple layers, each capturing different levels of representation. While prior studies explored their layer-wise representations for efficiency and performance, speech quality assessment (SQA) models predominantly rely on last-layer features, leaving intermediate layers underexamined. In this work, we systematically evaluate different layers of multiple SSL models for predicting mean-opinion-score (MOS). Features from each layer are fed into a lightweight regression network to assess effectiveness. Our experiments consistently show early-layers features outperform or match those from the last layer, leading to significant improvements over conventional approaches and state-of-the-art MOS prediction models. These findings highlight the advantages of early-layer selection, offering enhanced performance and reduced system complexity.

eess.AS↗

arXiv2025

Leveraging LLMs for Scalable Non-intrusive Speech Quality Assessment

Non-intrusive speech quality assessment (SQA) systems suffer from limited training data and costly human annotations, hindering their generalization to real-time conferencing calls. In this work, we propose leveraging large language models (LLMs) as pseudo-raters for speech quality to address these data bottlenecks. We construct LibriAugmented, a dataset consisting of 101,129 speech clips with simulated degradations labeled by a fine-tuned auditory LLM (Vicuna-7b-v1.5). We compare three training strategies: using human-labeled data, using LLM-labeled data, and a two-stage approach (pretraining on LLM labels, then fine-tuning on human labels), using both DNSMOS Pro and DeePMOS. We test on several datasets across languages and quality degradations. While LLM-labeled training yields mixed results compared to human-labeled training, we provide empirical evidence that the two-stage approach improves the generalization performance (e.g., DNSMOS Pro achieves 0.63 vs. 0.55 PCC on NISQA_TEST_LIVETALK and 0.73 vs. 0.65 PCC on Tencent with reverb). Our findings demonstrate the potential of using LLMs as scalable pseudo-raters for speech quality assessment, offering a cost-effective solution to the data limitation problem.

eess.AS↗

arXiv2025

S-RRG-Bench: Structured Radiology Report Generation with Fine-Grained Evaluation Framework

Radiology report generation (RRG) for diagnostic images, such as chest X-rays, plays a pivotal role in both clinical practice and AI. Traditional free-text reports suffer from redundancy and inconsistent language, complicating the extraction of critical clinical details. Structured radiology report generation (S-RRG) offers a promising solution by organizing information into standardized, concise formats. However, existing approaches often rely on classification or visual question answering (VQA) pipelines that require predefined label sets and produce only fragmented outputs. Template-based approaches, which generate reports by replacing keywords within fixed sentence patterns, further compromise expressiveness and often omit clinically important details. In this work, we present a novel approach to S-RRG that includes dataset construction, model training, and the introduction of a new evaluation framework. We first create a robust chest X-ray dataset (MIMIC-STRUC) that includes disease names, severity levels, probabilities, and anatomical locations, ensuring that the dataset is both clinically relevant and well-structured. We train an LLM-based model to generate standardized, high-quality reports. To assess the generated reports, we propose a specialized evaluation metric (S-Score) that not only measures disease prediction accuracy but also evaluates the precision of disease-specific details, thus offering a clinically meaningful metric for report quality that focuses on elements critical to clinical decision-making and demonstrates a stronger alignment with human assessments. Our approach highlights the effectiveness of structured reports and the importance of a tailored evaluation metric for S-RRG, providing a more clinically relevant measure of report quality.

cs.CV↗

arXiv2025

Multivariate Probabilistic Assessment of Speech Quality

The mean opinion score (MOS) is a standard metric for assessing speech quality, but its singular focus fails to identify specific distortions when low scores are observed. The NISQA dataset addresses this limitation by providing ratings across four additional dimensions: noisiness, coloration, discontinuity, and loudness, alongside MOS. In this paper, we extend the explored univariate MOS estimation to a multivariate framework by modeling these dimensions jointly using a multivariate Gaussian distribution. Our approach utilizes Cholesky decomposition to predict covariances without imposing restrictive assumptions and extends probabilistic affine transformations to a multivariate context. Experimental results show that our model performs on par with state-of-the-art methods in point estimation, while uniquely providing uncertainty and correlation estimates across speech quality dimensions. This enables better diagnosis of poor speech quality and informs targeted improvements.

eess.AS↗

arXiv2025

Human-in-the-Loop AI for HVAC Management Enhancing Comfort and Energy Efficiency

Heating, Ventilation, and Air Conditioning (HVAC) systems account for approximately 38% of building energy consumption globally, making them one of the most energy-intensive services. The increasing emphasis on energy efficiency and sustainability, combined with the need for enhanced occupant comfort, presents a significant challenge for traditional HVAC systems. These systems often fail to dynamically adjust to real-time changes in electricity market rates or individual comfort preferences, leading to increased energy costs and reduced comfort. In response, we propose a Human-in-the-Loop (HITL) Artificial Intelligence framework that optimizes HVAC performance by incorporating real-time user feedback and responding to fluctuating electricity prices. Unlike conventional systems that require predefined information about occupancy or comfort levels, our approach learns and adapts based on ongoing user input. By integrating the occupancy prediction model with reinforcement learning, the system improves operational efficiency and reduces energy costs in line with electricity market dynamics, thereby contributing to demand response initiatives. Through simulations, we demonstrate that our method achieves significant cost reductions compared to baseline approaches while maintaining or enhancing occupant comfort. This feedback-driven approach ensures personalized comfort control without the need for predefined settings, offering a scalable solution that balances individual preferences with economic and environmental goals.

eess.SY↗

arXiv2025

Impairments are Clustered in Latents of Deep Neural Network-based Speech Quality Models

In this article, we provide an experimental observation: Deep neural network (DNN) based speech quality assessment (SQA) models have inherent latent representations where many types of impairments are clustered. While DNN-based SQA models are not trained for impairment classification, our experiments show good impairment classification results in an appropriate SQA latent representation. We investigate the clustering of impairments using various kinds of audio degradations that include different types of noises, waveform clipping, gain transition, pitch shift, compression, reverberation, etc. To visualize the clusters we perform classification of impairments in the SQA-latent representation domain using a standard k-nearest neighbor (kNN) classifier. We also develop a new DNN-based SQA model, named DNSMOS+, to examine whether an improvement in SQA leads to an improvement in impairment classification. The classification accuracy is 94% for LibriAugmented dataset with 16 types of impairments and 54% for ESC-50 dataset with 50 types of real noises.

eess.AS↗

arXiv2025

Learning and Generating Diverse Residential Load Patterns Using GAN with Weakly-Supervised Training and Weight Selection

The scarcity of high-quality residential load data can pose obstacles for decarbonizing the residential sector as well as effective grid planning and operation. The above challenges have motivated research into generating synthetic load data, but existing methods faced limitations in terms of scalability, diversity, and similarity. This paper proposes a Generative Adversarial Network-based Synthetic Residential Load Pattern (RLP-GAN) generation model, a novel weakly-supervised GAN framework, leveraging an over-complete autoencoder to capture dependencies within complex and diverse load patterns and learn household-level data distribution at scale. We incorporate a model weight selection method to address the mode collapse problem and generate load patterns with high diversity. We develop a holistic evaluation method to validate the effectiveness of RLP-GAN using real-world data of 417 households. The results demonstrate that RLP-GAN outperforms state-of-the-art models in capturing temporal dependencies and generating load patterns with higher similarity to real data. Furthermore, we have publicly released the RLP-GAN generated synthetic dataset, which comprises one million synthetic residential load pattern profiles.

cs.LG↗

arXiv2025

ReFINE: A Reward-Based Framework for Interpretable and Nuanced Evaluation of Radiology Report Generation

Automated radiology report generation (R2Gen) has advanced significantly, introducing challenges in accurate evaluation due to its complexity. Traditional metrics often fall short by relying on rigid word-matching or focusing only on pathological entities, leading to inconsistencies with human assessments. To bridge this gap, we introduce ReFINE, an automatic evaluation metric designed specifically for R2Gen. Our metric utilizes a reward model, guided by our margin-based reward enforcement loss, along with a tailored training data design that enables customization of evaluation criteria to suit user-defined needs. It not only scores reports according to user-specified criteria but also provides detailed sub-scores, enhancing interpretability and allowing users to adjust the criteria between different aspects of reports. Leveraging GPT-4, we designed an easy-to-use data generation pipeline, enabling us to produce extensive training data based on two distinct scoring systems, each containing reports of varying quality along with corresponding scores. These GPT-generated reports are then paired as accepted and rejected samples through our pairing rule to train an LLM towards our fine-grained reward model, which assigns higher rewards to the report with high quality. Our reward-control loss enables this model to simultaneously output multiple individual rewards corresponding to the number of evaluation criteria, with their summation as our final ReFINE. Our experiments demonstrate ReFINE's heightened correlation with human judgments and superior performance in model selection compared to traditional metrics. Notably, our model provides both an overall score and individual scores for each evaluation item, enhancing interpretability. We also demonstrate its flexible training across various evaluation systems.

cs.CL↗

arXiv2024

Synthetic Data Generation for Residential Load Patterns via Recurrent GAN and Ensemble Method

Generating synthetic residential load data that can accurately represent actual electricity consumption patterns is crucial for effective power system planning and operation. The necessity for synthetic data is underscored by the inherent challenges associated with using real-world load data, such as privacy considerations and logistical complexities in large-scale data collection. In this work, we tackle the above-mentioned challenges by developing the Ensemble Recurrent Generative Adversarial Network (ERGAN) framework to generate high-fidelity synthetic residential load data. ERGAN leverages an ensemble of recurrent Generative Adversarial Networks, augmented by a loss function that concurrently takes into account adversarial loss and differences between statistical properties. Our developed ERGAN can capture diverse load patterns across various households, thereby enhancing the realism and diversity of the synthetic data generated. Comprehensive evaluations demonstrate that our method consistently outperforms established benchmarks in the synthetic generation of residential load data across various performance metrics including diversity, similarity, and statistical measures. The findings confirm the potential of ERGAN as an effective tool for energy applications requiring synthetic yet realistic load data. We also make the generated synthetic residential load patterns publicly available.

cs.LG↗

arXiv2024

MedXChat: A Unified Multimodal Large Language Model Framework towards CXRs Understanding and Generation

Multimodal Large Language Models (MLLMs) have shown success in various general image processing tasks, yet their application in medical imaging is nascent, lacking tailored models. This study investigates the potential of MLLMs in improving the understanding and generation of Chest X-Rays (CXRs). We introduce MedXChat, a unified framework facilitating seamless interactions between medical assistants and users for diverse CXR tasks, including text report generation, visual question-answering (VQA), and Text-to-CXR generation. Our MLLMs using natural language as the input breaks task boundaries, maximally simplifying medical professional training by allowing diverse tasks within a single environment. For CXR understanding, we leverage powerful off-the-shelf visual encoders (e.g., ViT) and LLMs (e.g., mPLUG-Owl) to convert medical imagery into language-like features, and subsequently fine-tune our large pre-trained models for medical applications using a visual adapter network and a delta-tuning approach. For CXR generation, we introduce an innovative synthesis approach that utilizes instruction-following capabilities within the Stable Diffusion (SD) architecture. This technique integrates smoothly with the existing model framework, requiring no extra parameters, thereby maintaining the SD's generative strength while also bestowing upon it the capacity to render fine-grained medical images with high fidelity. Through comprehensive experiments, our model demonstrates exceptional cross-task adaptability, displaying adeptness across all three defined tasks. Our MedXChat model and the instruction dataset utilized in this research will be made publicly available to encourage further exploration in the field.

cs.CV↗

arXiv2024

MRScore: Evaluating Radiology Report Generation with LLM-based Reward System

In recent years, automated radiology report generation has experienced significant growth. This paper introduces MRScore, an automatic evaluation metric tailored for radiology report generation by leveraging Large Language Models (LLMs). Conventional NLG (natural language generation) metrics like BLEU are inadequate for accurately assessing the generated radiology reports, as systematically demonstrated by our observations within this paper. To address this challenge, we collaborated with radiologists to develop a framework that guides LLMs for radiology report evaluation, ensuring alignment with human analysis. Our framework includes two key components: i) utilizing GPT to generate large amounts of training data, i.e., reports with different qualities, and ii) pairing GPT-generated reports as accepted and rejected samples and training LLMs to produce MRScore as the model reward. Our experiments demonstrate MRScore's higher correlation with human judgments and superior performance in model selection compared to traditional metrics. Our code and datasets will be available on GitHub.

cs.CL↗