arXiv subjects

Yang Cui

Publications and source records attributed to Yang Cui.

At least 19 recordsLinked to original sources

arXiv2026

SoftVTBench: A Deformation-Aware Visuo-Tactile Dataset and Benchmark for Deformable-Object Manipulation

Physical interaction quality is central to deformable-object manipulation, yet most benchmarks evaluate task success alone. A policy may complete the task while allowing slip or causing excessive compression. A primary bottleneck is the absence of visuo-tactile datasets that pair policy-visible contact observations with independent physical ground truth over complete tasks. We introduce SoftVTBench, a visuo-tactile dataset for physical-interaction-aware deformable-object manipulation. It contains 4,000 expert demonstrations and more than 50 assets, including volumetric deformable objects and visually matched rigid twins. At 20 Hz, each episode synchronizes multi-view RGB, dual-finger tactile RGB and marker motion, proprioception, language, and binary and continuous gripper actions, alongside evaluator-only finite-element (FEM) states. Building upon this dataset, we establish a closed-loop benchmark that uses fixed object-specific calibration to define the Deformation-aware Success Rate (DSR), which counts a rollout as successful only when it completes the task and keeps peak normalized deformation within tolerance. Across Diffusion Policy, $\pi_{0.5}$, and FastWAM, all 12 in-distribution configurations contain successful rollouts that violate the deformation tolerance, accounting for 0.7--24% of each configuration's successes. Under distribution shift, visuo-tactile variants achieve higher task success in all six policy--suite comparisons and higher DSR in five, whereas their in-distribution benefits are mixed. These results show that making touch available does not by itself ensure effective multimodal fusion. SoftVTBench therefore provides a common visuo-tactile resource for studying not only whether a policy succeeds, but how it physically interacts with deformable objects and when touch improves that interaction.

cs.RO↗

arXiv2026

TrajWiki: Source-Grounded Memory Trajectories for Long-Horizon Dialogue Agents

Large language model agents have shown strong capabilities in generating coherent and contextually appropriate responses, yet robust long-horizon dialogue remains limited by the lack of external memory that is traceable, updatable, and diagnostically transparent. Existing memory-augmented agents often store memories as isolated records or overwritable states, making it difficult to preserve how information originates, evolves, conflicts, or becomes obsolete over time. We propose TrajWiki, a trajectory-based memory framework for long-horizon conversational agents. Instead of treating memory as static entries, TrajWiki represents each memory as a source-grounded evolution trajectory, maintained through immutable episodic snapshots and claim-level operations such as ADD, REVISE, and DEPRECATE. To reduce fragmentation and retrieval cost, TrajWiki further introduces Memory Wiki, a persistent intermediate layer that incrementally compiles dialogue history into structured and interlinked wiki pages capturing salient entities, events, quantities, topics, and conflicts. At inference time, queries are routed hierarchically from relevant wiki pages to linked memory trajectories, then to corresponding snapshots and source messages for evidence-grounded answer synthesis. Experiments on LoCoMo and MedMT show that TrajWiki improves long-horizon dialogue performance across both open-source and closed-source LLM backbones, while providing greater interpretability and diagnostic visibility into memory evolution, retrieval failures, and answer generation.

cs.AI↗

arXiv2026

Dispersion-Guided Physics-Aware Deep Inverse Operator for Surface Wave Mode Separation

Surface-wave (SW) dispersion analysis is widely used in near-surface geophysics and seismology to determine shear-wave velocity structures by measuring SW geometric dispersion in seismic data. Among the available approaches, multichannel analysis of surface waves (MASW) and two-station methods are commonly employed to extract dispersion information for SW inversion. However, the coexistence of fundamental and higher modes in seismic data poses challenges for these methods, particularly for two-station analysis. To separate the different mode components, we propose a physics-aware unsupervised deep-learning framework. The method acts as a deep inverse operator that directly separates fundamental- and higher-mode components in the time-space domain using an adaptive Gaussian mask constructed in the frequency-phase-velocity (f-v) domain. Physical constraints are incorporated into the loss function by maximizing energy concentration within the target mask while suppressing leakage outside it. Through backpropagation, the network learns the inverse mapping from physical constraints in the f-v domain to wavefield separation in the time-space domain without requiring labeled training data. Numerical experiments on both synthetic and field data show that the framework provides a robust and automated solution for SW mode separation, facilitating more reliable dispersion-curve picking and improving the accuracy of subsequent SW inversion.

physics.geo-ph↗

arXiv2026

SoftVTBench: A Safety-Aware Visuo-Tactile Benchmark for Physically Constrained Robotic Manipulation of Deformable Objects (Early Version)

Deformable object manipulation poses challenges beyond task completion: successful execution must also maintain safe physical interaction, holding the object stably without slip or drop while avoiding excessive deformation. However, existing manipulation benchmarks are predominantly success-oriented and rarely evaluate whether a policy remains physically safe throughout execution. We present SoftVTBench, a safety-aware visuo-tactile benchmark for physically constrained deformable object manipulation. Built in Isaac Sim with finite-element-simulated deformable objects, SoftVTBench provides multi-view RGB observations, RGB tactile sensing with marker motion, proprioception, and language instructions, and defines four matched task suites over object type (deformable vs. rigid) and variation axis (object vs. spatial). It separately reports Goal Success and Safety Success; the latter additionally requires no drop and peak deformation below a calibrated object-specific threshold, measured from policy-hidden privileged Finite Element Method (FEM) states. We implement pi0.5-based baselines under this protocol. Experiments show that success-only evaluation substantially overstates policy performance, as a large fraction of goal-completing rollouts still violate physical safety. Furthermore, incorporating tactile sensing improves Safety Success (e.g., from 21.4% to 35.6% on object-centric deformable tasks) and reduces object deformation during execution, while maintaining comparable Goal Success. SoftVTBench provides a reproducible benchmark for studying visuo-tactile deformable manipulation under physical interaction constraints.

cs.RO↗

arXiv2026

Are Reasoning Vision-Language Models Robust to Semantic Visual Distractions?

Reasoning Vision-Language Models (VLMs) achieve strong performance on complex multimodal tasks, but reliable real-world application requires handling visual inputs that are messier than clean, curated benchmarks. Existing works mainly evaluate such reliability of VLMs through input corruptions, such as noise, blur and weather effects, which make visual evidence harder to perceive. This leaves a critical reliability failure mode underexplored: a model may perceive the evidence correctly, yet reason from plausible but irrelevant and distracting evidence and propagate this mistake to its final answer. To address this gap, we introduce \textbf{Distract-Bench}, a benchmark for evaluating VLM robustness to \textbf{semantic visual distractions}, defined as meaningful but task-irrelevant visual cues added to inputs while preserving the ground-truth answer. We comprehensively evaluate eight leading open-source and two closed-source VLMs across conventional vision corruptions and Distract-Bench. Our results show that Distract-Bench exposes a robustness failure distinct from vision corruptions: reasoning VLMs largely track their non-reasoning base models under perceptual degradation, but show consistently lower robustness to semantic distractions. Further analysis shows that these distractions often enter the reasoning process of VLMs, are treated as evidence, and lead to incorrect answers. Together, these findings reframe robustness evaluation for reasoning VLMs, shifting the focus from degraded perception to distractions for reliable real-world visual reasoning. Our data and code are available at https://github.com/Yizheng-Sun/Distract-Bench.

cs.CV↗

arXiv2026

Parameter-Efficient Adaptation of Pre-Trained Vision Foundation Models for Active and Passive Seismic Data Denoising

The demand for high-resolution subsurface imaging and continuous Earth monitoring has driven rapid growth in active and passive seismic data from dense geophone deployments, distributed acoustic sensing (DAS) arrays, and large-scale 2D and 3D surveys. This expansion makes complex noise suppression increasingly challenging, especially when signal fidelity must be preserved. Conventional supervised deep learning methods are often task-specific, require large paired datasets, and can suffer from domain shift under new acquisition conditions. Foundation models offer a promising alternative, but pre-training seismic foundation models from scratch requires massive domain-specific data and substantial computation. We propose an efficient framework that repurposes general-purpose Vision Foundation Models (VFMs) for geophysical tasks through Parameter-Efficient Fine-Tuning. The architecture uses a pre-trained VFM, a DINOv3 encoder, adapted with Low-Rank Adaptation (LoRA) to enable effective feature adaptation with few additional parameters. To improve robustness under unseen field conditions without ground truth, we introduce a kurtosis-guided unsupervised test-time adaptation module that updates only LoRA parameters during inference. This module self-calibrates the model to site-specific noise by identifying information-rich regions via kurtosis and performing self-training without labeled data. Experiments on public exploration seismic images and DAS vertical seismic profiling data from the Utah FORGE site show that the framework matches or outperforms domain-specific models. Tests on unseen cross-site data from a land survey in China and the Gro{\ss} Sch\"onebeck geothermal site in Germany further demonstrate strong generalization and effective signal-noise separation. These results highlight the potential of adapting pre-trained VFMs to data-intensive problems in exploration seismology.

physics.geo-ph↗

arXiv2026

Computational Lesions in Multilingual Language Models Separate Shared and Language-specific Brain Alignment

How the brain supports language across different languages is a basic question in neuroscience and a useful test for multilingual artificial intelligence. Neuroimaging has identified language-responsive brain regions across languages, but it cannot by itself show whether the underlying processing is shared or language-specific. Here we use six multilingual large language models (LLMs) as controllable systems and create targeted ``computational lesions'' by zeroing small parameter sets that are important across languages or especially important for one language. We then compare intact and lesioned models in predicting functional magnetic resonance imaging (fMRI) responses during 100 minutes of naturalistic story listening in native English, Chinese and French (112 participants). Lesioning a compact shared core reduces whole-brain encoding correlation by 60.32% relative to intact models, whereas language-specific lesions preserve cross-language separation in embedding space but selectively weaken brain predictivity for the matched native language. These results support a shared backbone with embedded specializations and provide a causal framework for studying multilingual brain-model alignment.

cs.CL↗

arXiv2025

OmniTFT: Omni Target Forecasting for Vital Signs and Laboratory Result Trajectories in Multi Center ICU Data

Accurate multivariate time-series prediction of vital signs and laboratory results is crucial for early intervention and precision medicine in intensive care units (ICUs). However, vital signs are often noisy and exhibit rapid fluctuations, while laboratory tests suffer from missing values, measurement lags, and device-specific bias, making integrative forecasting highly challenging. To address these issues, we propose OmniTFT, a deep learning framework that jointly learns and forecasts high-frequency vital signs and sparsely sampled laboratory results based on the Temporal Fusion Transformer (TFT). Specifically, OmniTFT implements four novel strategies to enhance performance: sliding window equalized sampling to balance physiological states, frequency-aware embedding shrinkage to stabilize rare-class representations, hierarchical variable selection to guide model attention toward informative feature clusters, and influence-aligned attention calibration to enhance robustness during abrupt physiological changes. By reducing the reliance on target-specific architectures and extensive feature engineering, OmniTFT enables unified modeling of multiple heterogeneous clinical targets while preserving cross-institutional generalizability. Across forecasting tasks, OmniTFT achieves substantial performance improvement for both vital signs and laboratory results on the MIMIC-III, MIMIC-IV, and eICU datasets. Its attention patterns are interpretable and consistent with known pathophysiology, underscoring its potential utility for quantitative decision support in clinical care.

cs.LG↗

arXiv2025

Learning from Imperfect Labels: A Physics-Aware Neural Operator with Application to DAS Data Denoising

Supervised deep learning methods typically require large datasets and high-quality labels to achieve reliable predictions. However, their performance often degrades when trained on imperfect labels. To address this challenge, we propose a physics-aware loss function that serves as a penalty term to mitigate label imperfections during training. In addition, we introduce a modified U-Net-Enhanced Fourier Neural Operator (UFNO) that achieves high-fidelity feature representation while leveraging the advantages of operator learning in function space. By combining these two components, we develop a physics-aware UFNO (PAUFNO) framework that effectively learns from imperfect labels. To evaluate the proposed framework, we apply it to the denoising of distributed acoustic sensing (DAS) data from the Utah FORGE site. The label data were generated using an integrated filtering-based method, but still contain residual coupling noise in the near-wellbore channels. The denoising workflow incorporates a patching-based data augmentation strategy, including an uplifting step, spatial-domain convolutional operations, spectral convolution, and a projection layer to restore data to the desired shape. Extensive numerical experiments demonstrate that the proposed framework achieves superior denoising performance, effectively enhancing DAS records and recovering hidden signals with high accuracy.

physics.geo-ph↗

arXiv2025

Efficient Speech Watermarking for Speech Synthesis via Progressive Knowledge Distillation

With the rapid advancement of speech generative models, unauthorized voice cloning poses significant privacy and security risks. Speech watermarking offers a viable solution for tracing sources and preventing misuse. Current watermarking technologies fall mainly into two categories: DSP-based methods and deep learning-based methods. DSP-based methods are efficient but vulnerable to attacks, whereas deep learning-based methods offer robust protection at the expense of significantly higher computational cost. To improve the computational efficiency and enhance the robustness, we propose PKDMark, a lightweight deep learning-based speech watermarking method that leverages progressive knowledge distillation (PKD). Our approach proceeds in two stages: (1) training a high-performance teacher model using an invertible neural network-based architecture, and (2) transferring the teacher's capabilities to a compact student model through progressive knowledge distillation. This process reduces computational costs by 93.6% while maintaining high level of robust performance and imperceptibility. Experimental results demonstrate that our distilled model achieves an average detection F1 score of 99.6% with a PESQ of 4.30 in advanced distortions, enabling efficient speech watermarking for real-time speech synthesis applications.

cs.SD↗

arXiv2025

Transformer Enhanced Relation Classification: A Comparative Analysis of Contextuality, Data Efficiency and Sequence Complexity

In the era of large language model, relation extraction (RE) plays an important role in information extraction through the transformation of unstructured raw text into structured data (Wadhwa et al., 2023). In this paper, we systematically compare the performance of deep supervised learning approaches without transformers and those with transformers. We used a series of non-transformer architectures such as PA-LSTM(Zhang et al., 2017), C-GCN(Zhang et al., 2018), and AGGCN(attention guide GCN)(Guo et al., 2019), and a series of transformer architectures such as BERT, RoBERTa, and R-BERT(Wu and He, 2019). Our comparison included traditional metrics like micro F1, as well as evaluations in different scenarios, varying sentence lengths, and different percentages of the dataset for training. Our experiments were conducted on TACRED, TACREV, and RE-TACRED. The results show that transformer-based models outperform non-transformer models, achieving micro F1 scores of 80-90% compared to 64-67% for non-transformer models. Additionally, we briefly review the research journey in supervised relation classification and discuss the role and current status of large language models (LLMs) in relation extraction.

cs.CL↗

arXiv2025

Transient Chirality in the Gelation of Adhesive Spinner Monolayers

Active systems of self-rotating elements inherently exhibit chirality, making them of fundamental interest due to parity violation. Using large-scale hydrodynamic simulations, we investigate the gelation of adhesive spinners confined to quasi-2D monolayers at low Reynolds numbers. Unlike the coarsening dynamics of passive colloids, spinner gelation follows a different pathway, displaying structural chirality during the early stages of aggregation. However, this chirality dissipates upon dynamical arrest, resulting in a final gel structure that resembles a conventional colloidal gel. As a result, we find no sign of odd mechanical responses. Nonetheless, the elastic modulus and gelation time remain tunable through spinning activity, providing a new avenue for the bottom-up design of programmable soft materials.

cond-mat.soft↗

arXiv2023

Deep Reinforcement Learning Based Optimal Energy Management of Multi-energy Microgrids with Uncertainties

Multi-energy microgrid (MEMG) offers an effective approach to deal with energy demand diversification and new energy consumption on the consumer side. In MEMG, it is critical to deploy an energy management system (EMS) for efficient utilization of energy and reliable operation of the system. To help EMS formulate optimal dispatching schemes, a deep reinforcement learning (DRL)-based MEMG energy management scheme with renewable energy source (RES) uncertainty is proposed in this paper. To accurately describe the operating state of the MEMG, the off-design performance model of energy conversion devices is considered in scheduling. The nonlinear optimal dispatching model is expressed as a Markov decision process (MDP) and is then addressed by the twin delayed deep deterministic policy gradient (TD3) algorithm. In addition, to accurately describe the uncertainty of RES, the conditional-least squares generative adversarial networks (C-LSGANs) method based on RES forecast power is proposed to construct the scenarios set of RES power generation. The generated data of RES is used for scheduling to obtain caps and floors for the purchase of electricity and natural gas. Based on this, the superior energy supply sector can formulate solutions in advance to tackle the uncertainty of RES. Finally, the simulation analysis demonstrates the validity and superiority of the method.

eess.SY↗

arXiv2023

Filled Colloidal Gel Rheology: Strengthening, Stiffening, and Tunability

Filler-induced strengthening is ubiquitous in materials science and is particularly well-established in polymeric nanocomposites. Despite having similar constituents, colloidal gels with solid filling exhibit distinct rheology, which is of practical interest to industry (e.g., lithium-ion batteries) yet remains poorly understood. We show, using experiments and simulations, that filling monotonically enhances the yield stress (i.e., strength) of colloidal gels while the elastic modulus (i.e., stiffness) first increases and then decreases. The latter softening effect results from a frustrated gel matrix at dense filling, evidenced by a growing inter-phase pressure. This structural frustration is, however, not detrimental to yielding resistance. Instead, fillers offer additional mechanical support to the gel backbone via percolating force chains, decreasing the yield strain at the same time. We develop a mechanistic picture of this phenomenology that leads us to a novel `filler-removal protocol,' making individual control over the strength and brittleness of a composite gel possible.

cond-mat.soft↗

arXiv2023

Short-term power load forecasting method based on CNN-SAEDN-Res

In deep learning, the load data with non-temporal factors are difficult to process by sequence models. This problem results in insufficient precision of the prediction. Therefore, a short-term load forecasting method based on convolutional neural network (CNN), self-attention encoder-decoder network (SAEDN) and residual-refinement (Res) is proposed. In this method, feature extraction module is composed of a two-dimensional convolutional neural network, which is used to mine the local correlation between data and obtain high-dimensional data features. The initial load fore-casting module consists of a self-attention encoder-decoder network and a feedforward neural network (FFN). The module utilizes self-attention mechanisms to encode high-dimensional features. This operation can obtain the global correlation between data. Therefore, the model is able to retain important information based on the coupling relationship between the data in data mixed with non-time series factors. Then, self-attention decoding is per-formed and the feedforward neural network is used to regression initial load. This paper introduces the residual mechanism to build the load optimization module. The module generates residual load values to optimize the initial load. The simulation results show that the proposed load forecasting method has advantages in terms of prediction accuracy and prediction stability.

eess.SP↗

arXiv2023

Collaborative Optimization of Multi-microgrids System with Shared Energy Storage Based on Multi-agent Stochastic Game and Reinforcement Learning

Achieving the economical and stable operation of Multi-microgrids (MMG) systems is vital. However, there are still some challenging problems to be solved. Firstly, from the perspective of stable operation, it is necessary to minimize the energy fluctuation of the main grid. Secondly, the characteristics of energy conversion equipment need to be considered. Finally, privacy protection while reducing the operating cost of an MMG system is crucial. To address these challenges, a Data-driven strategy for MMG systems with Shared Energy Storage (SES) is proposed. The Mixed-Attention is applied to fit the conditions of the equipment, additionally, Multi-Agent Soft Actor-Critic(MA-SAC) and (Multi-Agent Win or Learn Fast Policy Hill-Climbing)MA-WoLF-PHC are proposed to solve the partially observable dynamic stochastic game problem. By testing the operation data of the MMG system in Northwest China, following conclusions are drawn: the R-Square (R2) values of results reach 0.999, indicating the neural network effectively models the nonlinear conditions. The proposed MMG system framework can reduce energy fluctuations in the main grid by 1746.5kW in 24 hours and achieve a cost reduction of 16.21% in the test. Finally, the superiority of the proposed algorithms is verified through their fast convergence speed and excellent optimization performance.

eess.SY↗

arXiv2022

ResGrad: Residual Denoising Diffusion Probabilistic Models for Text to Speech

Denoising Diffusion Probabilistic Models (DDPMs) are emerging in text-to-speech (TTS) synthesis because of their strong capability of generating high-fidelity samples. However, their iterative refinement process in high-dimensional data space results in slow inference speed, which restricts their application in real-time systems. Previous works have explored speeding up by minimizing the number of inference steps but at the cost of sample quality. In this work, to improve the inference speed for DDPM-based TTS model while achieving high sample quality, we propose ResGrad, a lightweight diffusion model which learns to refine the output spectrogram of an existing TTS model (e.g., FastSpeech 2) by predicting the residual between the model output and the corresponding ground-truth speech. ResGrad has several advantages: 1) Compare with other acceleration methods for DDPM which need to synthesize speech from scratch, ResGrad reduces the complexity of task by changing the generation target from ground-truth mel-spectrogram to the residual, resulting into a more lightweight model and thus a smaller real-time factor. 2) ResGrad is employed in the inference process of the existing TTS model in a plug-and-play way, without re-training this model. We verify ResGrad on the single-speaker dataset LJSpeech and two more challenging datasets with multiple speakers (LibriTTS) and high sampling rate (VCTK). Experimental results show that in comparison with other speed-up methods of DDPMs: 1) ResGrad achieves better sample quality with the same inference speed measured by real-time factor; 2) with similar speech quality, ResGrad synthesizes speech faster than baseline methods by more than 10 times. Audio samples are available at https://resgrad1.github.io/.

eess.AS↗

arXiv2020

Electron Thermalization and Relaxation in Laser-Heated Nickel by Few-Femtosecond Core-Level Transient Absorption Spectroscopy

Direct measurements of photoexcited carrier dynamics in nickel are made using few-femtosecond extreme ultraviolet (XUV) transient absorption spectroscopy at the nickel M$_{2,3}$ edge. It is observed that the core-level absorption lineshape of photoexcited nickel can be described by a Gaussian broadening ($\sigma$) and a red shift ($\omega_{s}$) of the ground state absorption spectrum. Theory predicts, and the experimental results verify that after initial rapid carrier thermalization, the electron temperature increase ($\Delta T$) is linearly proportional to the Gaussian broadening factor $\sigma$, providing quantitative real-time tracking of the relaxation of the electron temperature. Measurements reveal an electron cooling time for 50 nm thick polycrystalline nickel films of 640$\pm$80 fs. With hot thermalized carriers, the spectral red shift exhibits a power-law relationship with the change in electron temperature of $\omega_{s}\propto\Delta T^{1.5}$. Rapid electron thermalization via carrier-carrier scattering accompanies and follows the nominal 4 fs photoexcitation pulse until the carriers reach a quasi-thermal equilibrium. Entwined with a <6 fs instrument response function, carrier thermalization times ranging from 34 fs to 13 fs are estimated from experimental data acquired at different pump fluences and it is observed that the electron thermalization time decreases with increasing pump fluence. The study provides an initial example of measuring electron temperature and thermalization in metals in real time with XUV light, and it lays a foundation for further investigation of photoinduced phase transitions and carrier transport in metals with core-level absorption spectroscopy.

cond-mat.mtrl-sci↗