arXiv subjects

Chen Xu

Publications and source records attributed to Chen Xu.

At least 73 records · Page 4Linked to original sources

arXiv2026

Exact Relation Between Wehrl-Rényi Entropy and Many-Body Entanglement

Quantum entanglement is key to understanding correlations and emergent phenomena in quantum many-body systems. For $N$ qubits (distinguishable spin-$1/2$ particles) in a pure quantum state, many-body entanglement can be characterized by the purity of the reduced density matrix of a subsystem, defined as the trace of the square of this reduced density matrix. Nevertheless, this approach depends on the choice of subsystem. In this letter, we establish an exact relation between the Wehrl-Rényi entropy (WRE) $S_W^{(2)}$, which is the 2nd Rényi entropy of the Husimi function of the entire system, and the purities of all possible subsystems. Specifically, we prove the relation $e^{-S_W^{(2)}} = (6π)^{-N} \sum_A \mathrm{Tr}({{\hat ρ}_A}^2)$, where $A$ denotes a subsystem with reduced density matrix ${\hat ρ}_A$, and the summation runs over all $2^N$ possible subsystems. Furthermore, we show that the WRE can be experimentally measured via a concrete scheme. Therefore, the WRE is a subsystem-independent and experimentally measurable characterization of the overall entanglement in pure states of $N$ qubits. It can be applied to the study of strongly correlated spin systems, particularly those with all-to-all couplings that do not have a natural subsystem division, such as systems realized with natural atoms in optical tweezer arrays or superconducting quantum circuits. We also analytically derive the WRE for several representative many-body states, including Haar-random states, the Greenberger-Horne-Zeilinger (GHZ) state, and the W state.

quant-ph↗

arXiv2026

All That Glisters Is Not Gold: A Benchmark for Reference-Free Counterfactual Financial Misinformation Detection

We introduce RFC Bench, a benchmark for evaluating large language models on financial misinformation under realistic news. RFC Bench operates at the paragraph level and captures the contextual complexity of financial news where meaning emerges from dispersed cues. The benchmark defines two complementary tasks: reference free misinformation detection and comparison based diagnosis using paired original perturbed inputs. Experiments reveal a consistent pattern: performance is substantially stronger when comparative context is available, while reference free settings expose significant weaknesses, including unstable predictions and elevated invalid outputs. These results indicate that current models struggle to maintain coherent belief states without external grounding. By highlighting this gap, RFC Bench provides a structured testbed for studying reference free reasoning and advancing more reliable financial misinformation detection in real world settings.

cs.CL↗

arXiv2026

Objective comparison of auditory profiles using manifold learning and intrinsic measures

Assigning individuals with hearing impairment to auditory profiles can support a better understanding of the causes and consequences of hearing loss and facilitate profile-based hearing-aid fitting. However, the factors influencing auditory profile generation remain insufficiently understood, and existing profiling frameworks have rarely been compared systematically. This study therefore investigated the impact of two key factors - the clustering method and the number of profiles - on auditory profile generation. In addition, eight established auditory profiling frameworks were systematically reviewed and compared using intrinsic statistical measures and manifold learning techniques. Frameworks were evaluated with respect to internal consistency (i.e., grouping similar individuals) and cluster separation (i.e., clear differentiation between groups). To ensure comparability, all analyses were conducted on a common open-access dataset, the extended Oldenburg Hearing Health Record (OHHR), comprising 1,127 participants (mean age = 67.2 years, SD = 12.0). Results showed that both the clustering method and the chosen number of profiles substantially influenced the resulting auditory profiles. Among purely audiogram-based approaches, the Bisgaard auditory profiles demonstrated the strongest clustering performance, whereas audiometric phenotypes performed worst. Among frameworks incorporating supra-threshold information in addition to the audiogram, the Hearing4All auditory profiles were advantageous, combining a near-optimal number of profile classes (N = 13) with high clustering quality, as indicated by a low Davies-Bouldin index. In conclusion, manifold learning and intrinsic measures enable systematic comparison of auditory profiling frameworks and identify the Hearing4All auditory profile as a promising approach for future research.

physics.med-ph↗

arXiv2025

Herman-Kluk-Like Semi-Classical Initial-Value Representation for Boltzmann Operator

The coherent-state initial-value representation (IVR) for the semi-classical real-time propagator of a quantum system, developed by Herman and Kluk (HK), is widely used in computational studies of chemical dynamics. On the other hand, the Boltzmann operator $e^{-\hat{H}/(k_B T)}$, with $\hat{H}$,$k_B$, and $T$ representing the Hamiltonian, Boltzmann constant, and temperature, respectively, plays a crucial role in chemical physics and other branches of quantum physics. One might naturally assume that a semi-classical IVR for the matrix element of this operator in the coordinate representation (i.e., $ \langle \tilde{x} | e^{-\hat{H}/(k_B T)} | x \rangle$, or the imaginary-time propagator) could be derived via a straightforward ``real-time $\rightarrow$ imaginary-time transformation'' from the HK IVR of the real-time propagator. However, this is not the case, as such a transformation results in a divergence in the high-temperature limit $(T \rightarrow \infty)$. In this work, we solve this problem and develop a reasonable HK-like semi-classical IVR for $ \langle \tilde{x} | e^{-\hat{H}/(k_B T)} | x \rangle$ specifically for systems where either the gradient of the potential energy (i.e., the force intensity) has a finite upper bound, or the potential becomes harmonic in the long-range limit. The integrand in this IVR is a real Gaussian function of the positions $x$ and $\tilde{x}$, which facilitates its application to realistic problems. Our HK-like IVR is exact for free particles and harmonic oscillators, and its effectiveness for other systems is demonstrated through numerical examples.

physics.chem-ph↗

arXiv2025

MFGDiffusion: Mask-Guided Smoke Synthesis for Enhanced Forest Fire Detection

Smoke is the first visible indicator of a wildfire.With the advancement of deep learning, image-based smoke detection has become a crucial method for detecting and preventing forest fires. However, the scarcity of smoke image data from forest fires is one of the significant factors hindering the detection of forest fire smoke. Image generation models offer a promising solution for synthesizing realistic smoke images. However, current inpainting models exhibit limitations in generating high-quality smoke representations, particularly manifesting as inconsistencies between synthesized smoke and background contexts. To solve these problems, we proposed a comprehensive framework for generating forest fire smoke images. Firstly, we employed the pre-trained segmentation model and the multimodal model to obtain smoke masks and image captions.Then, to address the insufficient utilization of masks and masked images by inpainting models, we introduced a network architecture guided by mask and masked image features. We also proposed a new loss function, the mask random difference loss, which enhances the consistency of the generated effects around the mask by randomly expanding and eroding the mask edges.Finally, to generate a smoke image dataset using random masks for subsequent detection tasks, we incorporated smoke characteristics and use a multimodal large language model as a filtering tool to select diverse and reasonable smoke images, thereby improving the quality of the synthetic dataset. Experiments showed that our generated smoke images are realistic and diverse, and effectively enhance the performance of forest fire smoke detection models. Code is available at https://github.com/wghr123/MFGDiffusion.

cs.CV↗

arXiv2025

Standard audiogram classification from loudness scaling data using unsupervised, supervised, and explainable machine learning techniques

To address the calibration and procedural challenges inherent in remote audiogram assessment for rehabilitative audiology, this study investigated whether calibration-independent adaptive categorical loudness scaling (ACALOS) data can be used to approximate individual audiograms by classifying listeners into standard Bisgaard audiogram types using machine learning. Three classes of machine learning approaches - unsupervised, supervised, and explainable - were evaluated. Principal component analysis (PCA) was performed to extract the first two principal components, which together explained more than 50 percent of the variance. Seven supervised multi-class classifiers were trained and compared, alongside unsupervised and explainable methods. Model development and evaluation used a large auditory reference database containing ACALOS data (N = 847). The PCA factor map showed substantial overlap between listeners, indicating that cleanly separating participants into six Bisgaard classes based solely on their loudness patterns is challenging. Nevertheless, the models demonstrated reasonable classification performance, with logistic regression achieving the highest accuracy among supervised approaches. These findings demonstrate that machine learning models can predict standard Bisgaard audiogram types, within certain limits, from calibration-independent loudness perception data, supporting potential applications in remote or resource-limited settings without requiring a traditional audiogram.

cs.SD↗

arXiv2025

Field-Tunable Anisotropic Fulde-Ferrell Phase in NbSe$_2$/CrSiTe$_3$ Heterostructures

The emergence of superconductivity in two-dimensional transition metal dichalcogenides with strong spin orbit coupling (SOC) has opened new avenues for exploring exotic superconducting states. Here, we report experimental observation of an anisotropic Fulde-Ferrell (FF) phase in few-layer NbSe$_2$/CrSiTe$_3$ heterostructures under in-plane magnetic fields. Through combined magnetoresistance and nonreciprocal transport measurements, we find that due to the couplings from the ferromagnetic CrSiTe$_3$, a half-dome-shaped region emerges in the magnetic field-temperature ($B$-$T$) diagram. Importantly, the half-dome-shaped region exhibits finite second harmonic resistance with in-plane anisotropy, indicating that the superconducting state is an anisotropic FF phase. Through a symmetry analysis combined with mean field calculations, we attribute the emergent anisotropic FF phase to the CrSiTe$_3$ layer induced Rashba SOC and three-fold rotational symmetry breaking. These results demonstrate that heterostructure stacking is a powerful tool for symmetry engineering in superconductors, which can advance the design of quantum devices in atomically thin superconducting materials.

cond-mat.supr-con↗

arXiv2025

BLM$_1$: A Boundless Large Model for Cross-Space, Cross-Task, and Cross-Embodiment Learning

Multimodal large language models (MLLMs) have advanced vision-language reasoning and are increasingly deployed in embodied agents. However, significant limitations remain: MLLMs generalize poorly across digital-physical spaces and embodiments; vision-language-action models (VLAs) produce low-level actions yet lack robust high-level embodied reasoning; and most embodied large language models (ELLMs) are constrained to digital-space with poor generalization to the physical world. Thus, unified models that operate seamlessly across digital and physical spaces while generalizing across embodiments and tasks remain absent. We introduce the \textbf{Boundless Large Model (BLM$_1$)}, a multimodal spatial foundation model that preserves instruction following and reasoning, incorporates embodied knowledge, and supports robust cross-embodiment control. BLM$_1$ integrates three key capabilities -- \textit{cross-space transfer, cross-task learning, and cross-embodiment generalization} -- via a two-stage training paradigm. Stage I injects embodied knowledge into the MLLM through curated digital corpora while maintaining language competence. Stage II trains a policy module through an intent-bridging interface that extracts high-level semantics from the MLLM to guide control, without fine-tuning the MLLM backbone. This process is supported by a self-collected cross-embodiment demonstration suite spanning four robot embodiments and six progressively challenging tasks. Evaluations across digital and physical benchmarks show that a single BLM$_1$ instance outperforms four model families -- MLLMs, ELLMs, VLAs, and GMLMs -- achieving $\sim\!\textbf{6%}$ gains in digital tasks and $\sim\!\textbf{3%}$ in physical tasks.

cs.AI↗

arXiv2025

UniPlanner: A Unified Motion Planning Framework for Autonomous Vehicle Decision-Making Systems via Multi-Dataset Integration

Motion planning is a critical component of autonomous vehicle decision-making systems, directly determining trajectory safety and driving efficiency. While deep learning approaches have advanced planning capabilities, existing methods remain confined to single-dataset training, limiting their robustness in planning. Through systematic analysis, we discover that vehicular trajectory distributions and history-future correlations demonstrate remarkable consistency across different datasets. Based on these findings, we propose UniPlanner, the first planning framework designed for multi-dataset integration in autonomous vehicle decision-making. UniPlanner achieves unified cross-dataset learning through three synergistic innovations. First, the History-Future Trajectory Dictionary Network (HFTDN) aggregates history-future trajectory pairs from multiple datasets, using historical trajectory similarity to retrieve relevant futures and generate cross-dataset planning guidance. Second, the Gradient-Free Trajectory Mapper (GFTM) learns robust history-future correlations from multiple datasets, transforming historical trajectories into universal planning priors. Its gradient-free design ensures the introduction of valuable priors while preventing shortcut learning, making the planning knowledge safely transferable. Third, the Sparse-to-Dense (S2D) paradigm implements adaptive dropout to selectively suppress planning priors during training for robust learning, while enabling full prior utilization during inference to maximize planning performance.

cs.AI↗

arXiv2025

AG-Fusion: adaptive gated multimodal fusion for 3d object detection in complex scenes

Multimodal camera-LiDAR fusion technology has found extensive application in 3D object detection, demonstrating encouraging performance. However, existing methods exhibit significant performance degradation in challenging scenarios characterized by sensor degradation or environmental disturbances. We propose a novel Adaptive Gated Fusion (AG-Fusion) approach that selectively integrates cross-modal knowledge by identifying reliable patterns for robust detection in complex scenes. Specifically, we first project features from each modality into a unified BEV space and enhance them using a window-based attention mechanism. Subsequently, an adaptive gated fusion module based on cross-modal attention is designed to integrate these features into reliable BEV representations robust to challenging environments. Furthermore, we construct a new dataset named Excavator3D (E3D) focusing on challenging excavator operation scenarios to benchmark performance in complex conditions. Our method not only achieves competitive performance on the standard KITTI dataset with 93.92% accuracy, but also significantly outperforms the baseline by 24.88% on the challenging E3D dataset, demonstrating superior robustness to unreliable modal information in complex industrial scenes.

cs.CV↗

arXiv2025

Unifying Search and Recommendation with Dual-View Representation Learning in a Generative Paradigm

Recommender systems and search engines serve as foundational elements of online platforms, with the former delivering information proactively and the latter enabling users to seek information actively. Unifying both tasks in a shared model is promising since it can enhance user modeling and item understanding. Previous approaches mainly follow a discriminative paradigm, utilizing shared encoders to process input features and task-specific heads to perform each task. However, this paradigm encounters two key challenges: gradient conflict and manual design complexity. From the information theory perspective, these challenges potentially both stem from the same issue -- low mutual information between the input features and task-specific outputs during the optimization process. To tackle these issues, we propose GenSR, a novel generative paradigm for unifying search and recommendation (S&R), which leverages task-specific prompts to partition the model's parameter space into subspaces, thereby enhancing mutual information. To construct effective subspaces for each task, GenSR first prepares informative representations for each subspace and then optimizes both subspaces in one unified model. Specifically, GenSR consists of two main modules: (1) Dual Representation Learning, which independently models collaborative and semantic historical information to derive expressive item representations; and (2) S&R Task Unifying, which utilizes contrastive learning together with instruction tuning to generate task-specific outputs effectively. Extensive experiments on two public datasets show GenSR outperforms state-of-the-art methods across S&R tasks. Our work introduces a new generative paradigm compared with previous discriminative methods and establishes its superiority from the mutual information perspective.

cs.IR↗

arXiv2025

Enhancing Speech Large Language Models with Prompt-Aware Mixture of Audio Encoders

Connecting audio encoders with large language models (LLMs) allows the LLM to perform various audio understanding tasks, such as automatic speech recognition (ASR) and audio captioning (AC). Most research focuses on training an adapter layer to generate a unified audio feature for the LLM. However, different tasks may require distinct features that emphasize either semantic or acoustic aspects, making task-specific audio features more desirable. In this paper, we propose Prompt-aware Mixture (PaM) to enhance the Speech LLM that uses multiple audio encoders. Our approach involves using different experts to extract different features based on the prompt that indicates different tasks. Experiments demonstrate that with PaM, only one Speech LLM surpasses the best performances achieved by all single-encoder Speech LLMs on ASR, Speaker Number Verification, and AC tasks. PaM also outperforms other feature fusion baselines, such as concatenation and averaging. Our code would be available at: https://github.com/shanweiqiao/PaM

eess.AS↗

arXiv2025

Siren's Song in the AI Ocean: A Survey on Hallucination in Large Language Models

While large language models (LLMs) have demonstrated remarkable capabilities across a range of downstream tasks, a significant concern revolves around their propensity to exhibit hallucinations: LLMs occasionally generate content that diverges from the user input, contradicts previously generated context, or misaligns with established world knowledge. This phenomenon poses a substantial challenge to the reliability of LLMs in real-world scenarios. In this paper, we survey recent efforts on the detection, explanation, and mitigation of hallucination, with an emphasis on the unique challenges posed by LLMs. We present taxonomies of the LLM hallucination phenomena and evaluation benchmarks, analyze existing approaches aiming at mitigating LLM hallucination, and discuss potential directions for future research.

cs.CL↗

arXiv2025

Hierarchical Multi-Interest Co-Network For Coarse-Grained Ranking

In this era of information explosion, a personalized recommendation system is convenient for users to get information they are interested in. To deal with billions of users and items, large-scale online recommendation services usually consist of three stages: candidate generation, coarse-grained ranking, and fine-grained ranking. The success of each stage depends on whether the model accurately captures the interests of users, which are usually hidden in users' behavior data. Previous research shows that users' interests are diverse, and one vector is not sufficient to capture users' different preferences. Therefore, many methods use multiple vectors to encode users' interests. However, there are two unsolved problems: (1) The similarity of different vectors in existing methods is too high, with too much redundant information. Consequently, the interests of users are not fully represented. (2) Existing methods model the long-term and short-term behaviors together, ignoring the differences between them. This paper proposes a Hierarchical Multi-Interest Co-Network (HCN) to capture users' diverse interests in the coarse-grained ranking stage. Specifically, we design a hierarchical multi-interest extraction layer to update users' diverse interest centers iteratively. The multiple embedded vectors obtained in this way contain more information and represent the interests of users better in various aspects. Furthermore, we develop a Co-Interest Network to integrate users' long-term and short-term interests. Experiments on several real-world datasets and one large-scale industrial dataset show that HCN effectively outperforms the state-of-the-art methods. We deploy HCN into a large-scale real world E-commerce system and achieve extra 2.5\% improvements on GMV (Gross Merchandise Value).

cs.IR↗

arXiv2025

Improving atomic force microscopy structure discovery via style-translation

Atomic force microscopy (AFM) is a key tool for characterising nanoscale structures, with functionalised tips now offering detailed images of the atomic structure. In parallel, AFM simulations using the particle probe model provide a cost-effective approach for rapid AFM image generation. Using state-of-the-art machine learning models and substantial simulated datasets, properties such as molecular structure, electrostatic potential, and molecular graph can be predicted from AFM images. However, transferring model performance from simulated to experimental AFM images poses challenges due to the subtle variations in real experimental data compared to the seemingly flawless simulations. In this study, we explore style translation to augment simulated images and improve the predictive performance of machine learning models in surface property analysis. We reduce the style gap between simulated and experimental AFM images and demonstrate the method's effectiveness in enhancing structure discovery models through local structural property distribution comparisons. This research presents a novel approach to improving the efficiency of machine learning models in the absence of labelled experimental data.

cond-mat.mtrl-sci↗

arXiv2025

Step-Audio 2 Technical Report

This paper presents Step-Audio 2, an end-to-end multi-modal large language model designed for industry-strength audio understanding and speech conversation. By integrating a latent audio encoder and reasoning-centric reinforcement learning (RL), Step-Audio 2 achieves promising performance in automatic speech recognition (ASR) and audio understanding. To facilitate genuine end-to-end speech conversation, Step-Audio 2 incorporates the generation of discrete audio tokens into language modeling, significantly enhancing its responsiveness to paralinguistic information such as speaking styles and emotions. To effectively leverage the rich textual and acoustic knowledge in real-world data, Step-Audio 2 integrates retrieval-augmented generation (RAG) and is able to call external tools such as web search to mitigate hallucination and audio search to switch timbres. Trained on millions of hours of speech and audio data, Step-Audio 2 delivers intelligence and expressiveness across diverse conversational scenarios. Evaluation results demonstrate that Step-Audio 2 achieves state-of-the-art performance on various audio understanding and conversational benchmarks compared to other open-source and commercial solutions. Please visit https://github.com/stepfun-ai/Step-Audio2 for more information.

cs.CL↗

arXiv2025

DLLMQuant: Quantizing Diffusion-based Large Language Models

Diffusion-based large language models (DLLMs) have shown promise for non-autoregressive text generation, but their deployment is constrained by large model sizes and heavy computational costs. Post-training quantization (PTQ), a widely used method for compressing and accelerating Large Language Models (LLMs), suffers from severe accuracy degradation and reduced generalization performance when directly applied to DLLMs (e.g., AWQ suffers a 16% accuracy drop on LLADA under W4A4). This paper explores how DLLMs' key mechanisms - dynamic masking, iterative generation, bidirectional attention - clash with quantization. We identify three core issues: 1) Iterative generation and dynamic masking ratios lead to distinct token distributions across decoding steps, which are not adequately captured by existing PTQ calibration methods; 2) Quantization errors are accumulated and amplified progressively during iteration in DLLMs, causing quantized models to perform worse as decoding steps progress; 3) Unmasked tokens stabilize while masked remain probabilistic, making overall feature distribution incompatible with existing PTQ methods. To address these issues, we propose DLLMQuant, a PTQ framework tailored for DLLMs, which incorporates three novel techniques: 1) Temporal-Mask Adaptive Sampling (TMAS), a calibration method that accounts for both time and mask factors, with the capacity to capture distributions across timesteps. 2) Interaction-Aware Activation Quantization (IA-AQ), which utilizes bidirectional attention's interaction signals to dynamically allocate quantization resources. 3) Certainty-Guided Quantization (CGQ), which integrates mask status and token scores as key weighting criteria into error compensation, making weight quantization more suitable for DLLMs. Experiments show that DLLMQuant achieves significant performance gains while enhancing efficiency.

cs.CL↗

arXiv2025

Calibration offset estimation in mobile hearing tests via categorical loudness scaling

Objective: To enable reliable smartphone-based hearing assessments by developing methods to estimate device calibration offsets using categorical loudness scaling (CLS). Design: Calibration offsets were simulated from a Gaussian distribution. Two prediction models - a Bayesian regression model and a nearest neighbor model - were trained on CLS-derived parameters and data from the Oldenburg Hearing Health Repository (OHHR). CLS was chosen because it provides level-independent measures (e.g., dynamic range) that remain robust despite calibration errors. Study Sample: The dataset comprised CLS results from N = 847 participants with a mean age of 70.0 years (SD = 8.7), including 556 male and 291 female listeners with diverse hearing profiles. Results: The Bayesian regression model achieved correlations of up to 0.81 between estimated and true calibration offsets, enabling accurate individual-level correction. Compared to threshold-based approaches, calibration uncertainty was reduced by factors between 0.41 and 0.79, demonstrating greater robustness in uncontrolled environments. Conclusions: CLS-based models can effectively compensate for missing calibration in mobile hearing assessments. This approach provides a practical alternative to threshold-based methods, supporting the use of smartphone-based tests outside laboratory settings and expanding access to reliable hearing healthcare in everyday and resource-limited contexts.

physics.med-ph↗