arXiv subjects

Qian Wang

Publications and source records attributed to Qian Wang.

At least 19 recordsLinked to original sources

arXiv2026

X2-NativeCursor: Native-Token Text Progress Tracking for Incremental-Text Streaming Codec TTS

Incremental-text streaming text-to-speech (TTS) needs online text progress tracking for synchronized highlighting, interruption handling, and dialogue-history updates. Input text arrives before it is spoken, so text arrival alone cannot indicate speech progress. Existing waveform-based alignment requires complete audio or adds acoustic processing during streaming. We propose X2-NativeCursor, a lightweight observer that tracks progress from native speech tokens before waveform decoding without changing the TTS generator. Its normalization plan links spoken labels to their original-text spans. Text and native-token encoders feed a local matcher that estimates the current label position. A separate output rule converts revisable position estimates into a cursor that never moves backward. Mean absolute error against an automatic reference is 0.151 Chinese characters with 80-ms lookahead, versus 1.253 characters with 320-ms lookahead for an online waveform baseline. Alignment real-time factor also decreases from 0.3598 to 0.0180 relative to this baseline. Lower tracking error is retained under a second automatic alignment reference. We evaluate X2-NativeCursor on Qwen3-TTS and validate its adaptation to CosyVoice2 by training a separate observer for each backbone. Code is publicly available at https://github.com/X-Square-Robot/X2Streaming-TTS.

cs.CL↗

arXiv2026

X2Streaming-ASR: wait when uncertain, emit when ready for streaming ASR

Streaming automatic speech recognition (ASR) for real-time voice agents and full-duplex dialogue must provide accurate partial transcripts with low commit latency. Existing systems commonly use a fixed chunk size, look-ahead, or target delay, or encourage emissions near estimated acoustic boundaries. These approaches do not directly optimize how much additional context to use at each output position under a single-pass, hard-commit constraint. We propose X2Streaming-ASR, which decomposes streaming recognition into when to commit and what to commit. Its three-stage training procedure first establishes streaming recognition ability, then warm-starts the commit policy with automatically probed trajectories, and finally refines the policy using character-level, segment-assigned group-relative rewards for recognition accuracy and latency. Across AISHELL-1/2/3 and WenetSpeech, X2Streaming-ASR achieves a mean character-level commit latency of 27-84 ms relative to forced-aligned character endpoints, compared with 409-585 ms for the evaluated streaming baselines. It achieves the best streaming CER among the evaluated systems on AISHELL-1 and AISHELL-3 with substantially lower latency.

cs.SD↗

arXiv2026

AutoUVM: Automated Prefetching Framework for LLMs under UVM Oversubscription

Large language models (LLMs) increasingly exceed the memory capacity of commodity GPUs, making memory oversubscription common in practical deployments. NVIDIA Unified Virtual Memory (UVM) provides transparent access to host memory, but its page-fault-driven migrations introduce severe performance overhead. While UVM exposes primitives (e.g., prefetching and placement hints) to mitigate these costs, they require low-level CUDA modifications, limiting their applicability for most LLM users. Meanwhile, existing UVM optimizations operate at coarse managed-object granularity and fail to capture deep learning frameworks' internal tensor-level memory behavior, leading to excessive data movement and CPU-GPU interconnect bottlenecks. We propose AutoUVM, an automated, framework-aware UVM prefetching system for efficient LLM execution under memory oversubscription. AutoUVM bridges the semantic gap between deep learning frameworks and UVM by exposing tensor-level access information and enabling policy-driven prefetching at fine granularity. Implemented as a transparent extension, AutoUVM requires no changes to model code and dynamically adapts to runtime memory pressure. We instantiate AutoUVM with a roofline-inspired policy to identify performance-critical data transfers. Across ten LLMs, AutoUVM achieves an average 3.1x speedup over baseline UVM and consistently surpasses the best-performing prior UVM prefetcher by 1.9x, with improvements of up to 4.7x over object-level prefetchers, while significantly reducing page faults.

cs.OS↗

arXiv2026

DualDiff3D: Dual Structure-Appearance Diffusion Priors for Reliability-Enhanced 3D Gaussian Splatting

While 3D Gaussian Splatting (3DGS) has revolutionized 3D reconstruction and novel-view synthesis, scenarios with limited input views often lead to poor reconstruction quality and artifacts in rendered novel views. Recent efforts attempt to utilize powerful diffusion priors, yet they typically process rendered and reference views concatenated along an additional dimension in a single network. These methods overlook an inherent nature that different views should maintain appearance similarity but differ in structure due to view shifts, leading to blur caused by conflicts between the two properties. In this paper, we propose DualDiff, a novel pipeline that leverages dual diffusion priors with a Structure-Appearance Attention (SAA) module to introduce reference guidance for refining low-quality novel views rendered from flawed 3D representations. Specifically, we retain one diffusion branch to focus on extracting structural information from the low-quality novel views, while introducing another branch to ensure appearance consistency with reference views. Furthermore, we present a 3D reconstruction framework named DualDiff3D, which integrates a reliability-enhanced Render-Refine-Optimize (RRO) loop to progressively and robustly incorporate the refined novel views, yielding more accurate 3DGS. Extensive experiments demonstrate that our approach outperforms state-of-the-art methods even in the inference-only setting, with further performance gains achievable through training. Our code and pre-trained weights are available at https://github.com/Akaneqwq/DualDiff3D.

cs.CV↗

arXiv2026

JudgeStealer: Extracting LLM Judging Capabilities across Evaluation Protocols

Large language model (LLM) judges are increasingly used across various evaluation scenarios, making their judgment capabilities valuable intellectual property. However, black-box access exposes these capabilities to model extraction attacks. Existing extraction methods do not specifically target LLM judges and provide limited support for multiple evaluation protocols under restricted query budgets. In this study, we propose JUDGESTEALER, the first query-efficient model extraction framework for replicating judging capabilities across pointwise scoring, pairwise comparison, and listwise ranking protocols. JUDGESTEALER exploits the strong cross-protocol agreement to acquire pointwise scores and transform them into pairwise and listwise supervisions without additional victim queries. To capture informative judge patterns and improve query efficiency, JUDGESTEALER dynamically selects pointwise inputs based on semantic diversity, predictive uncertainty, and potential judge biases. It further applies score smoothing and multi-protocol review to preserve the ordinal structure of scores and mitigate catastrophic forgetting during surrogate adaptation. Extensive experiments on state-of-the-art LLM-as-a-judge and reward models show that JUDGESTEALER consistently outperforms existing extraction baselines, achieving up to 73.3%, 87.0%, and 71.6% accuracy for pointwise, pairwise, and listwise evaluation, respectively. JUDGESTEALER also remains effective across different sur- rogate model scales, adaptation strategies, and reasoning settings. Moreover, JUDGESTEALER demonstrates robustness against representative extraction defenses.

cs.CL↗

arXiv2026

Machine Learning Unveils Finite-volume Energy Shifts in Three-body System

Finite-volume extrapolation (FVE) is essential for extracting physical observables in the lattice calculation. While rigorous FVE formulations are well established for short-range potentials in both two- and three-body systems, long-range interactions with force ranges comparable to the lattice size $L$ remain challenging. Extending a previous data-driven scheme for two-body systems, we apply symbolic regression (PySR) to uncover universal three-body FVE formulae. For short-range potentials, we reproduce the two limiting cases, i.e. $\kappa_3\gg\kappa_2$ and $\kappa_3\sim\kappa_2$. For pure long-range potentials, we obtain a dedicated analytic expression, and after incorporating short-range contributions, we uncover a unified formula consistent with the original PySR solution, which performs excellently in the intermediate force range around 1 fm. This work demonstrates that combining machine learning with physical constraints can yield novel analytical results inaccessible to conventional theoretical tools, advancing data-driven methodologies in hadron physics.

hep-lat↗

arXiv2026

WALL-SS: Scaling Long-horizon World Models via Next-Scale Autoregression

Generative world models provide robots with predictive models of how the world evolves under interaction, with growing potential for simulation, planning, policy evaluation, and robot learning. Beyond clip-level future prediction, a unified generative formulation should relate actions to consequences, support flexible horizons and continuous interaction, and enable reward-driven optimization. We introduce WALL-SS, a world model that generates visual futures through Scale-wise autoregressive Scaling, enabling action-controllable and long-horizon robotic simulation. WALL-SS represents embodied trajectories as causal sequences of temporally interleaved observations and actions, making action-dependent state transitions explicit while naturally supporting variable-length generation, streaming extension through reusable causal states, and direct optimization through sequence probabilities. To make this formulation effective over long horizons, we generate each future observation in a coarse-to-fine manner and develop three complementary components within the same hierarchy. Action-conditioned next-scale prediction injects scale-aligned action representations to improve action-future coupling and model both successful and failed behaviors. Scale-compressed long-horizon memory retains recent interactions at fine resolution while compressing distant observations and actions, with scale-wise dream forcing enhancing robustness to self-generated context. Finally, on-policy alignment optimizes autoregressive visual dynamics with action-following and long-term consistency rewards while preserving the pretrained visual distribution. Experiments show that WALL-SS improves action following and trajectory accuracy, supports coherent minute-long streaming rollout under bounded memory, and consistently benefits from on-policy alignment in reducing action drift and long-horizon inconsistency.

cs.RO↗

arXiv2026

Thermal width shift of $\Delta^{++}$ in a pion gas

We compute the thermal width shift of the $\Delta^{++}$ resonance induced by a pion gas within a nonrelativistic effective field theory framework. The $\Delta^{++}$ self-energy is evaluated from pion-forward-scattering diagrams with intermediate proton and $\Delta$ states, weighted by the thermal pion distribution. Analytical expressions for the imaginary part of the self-energy yield the temperature-dependent width correction $\delta\Gamma(T)$. The width increases with temperature, reaching approximately $6$~MeV at $T \approx 160$~MeV. When the temperature-dependent $\Delta$ and nucleon masses from an NJL-model chiral restoration scenario are incorporated, the width shift becomes non-monotonic, peaking near $T \approx 140$~MeV---a consequence of the competition between collisional broadening and the shrinking $\Delta \to N\pi$ phase space as the $N$--$\Delta$ mass gap closes. Applying our formalism to STAR data for $\Delta^{++}$ in d+Au collisions at $\sqrt{s_{NN}} = 200$~GeV, we extract a temperature $T \approx 300$~MeV at $p_t = 900$~MeV, consistent with the experimental extraction within errors. This value significantly exceeds the hadronic-phase temperature, explicitly demonstrating that the observed $\Delta^{++}$ width shift is not solely of pion-gas origin---genuine hot-medium and collective-flow contributions must be substantial.

hep-ph↗

arXiv2026

The Claws in Plain Sight: Unauthorized Context Disclosure through LLM Agent Tool Calls

LLM agents routinely construct tool-call arguments from user profiles, conversation history, retrieved documents, and prior tool results. However, legitimate access to contextual information does not imply authorization to transmit that information for every purpose or destination. We present Claw in Plain Sight, an authority- pressure attack in which task-adjacent content frames protected attributes as operationally or procedurally required, causing a model to include them in otherwise valid generated arguments. We evaluate Claw in Plain Sight using a controlled synthetic benchmark that crosses six pressure levels with four privacy-policy levels across five DeepSeek and Claude model configurations, producing 120 calls. Across the complete pressure-policy matrix, session-level disclosure rates range from 20.8% to 75.0% among the tested models. Stronger privacy instructions reduce aggregate disclosure but do not eliminate it consistently across models, showing that prompt-level policies do not provide a portable enforcement boundary. Our experiments use only synthetic profiles and capture proposed arguments locally; they measure policy-violating generation at the context-to-argument boundary, not completed network exfiltration or leakage from deployed users. These findings motivate purpose- and destination-aware inspection of generated tool arguments before execution.

cs.CR↗

arXiv2026

X2Streaming-TTS: Causal Token-Level Text-to-Speech from Streaming Text with Speech-State Inheritance

Streaming text-to-speech is essential for low-latency spoken dialogue systems, yet many systems wait for sentence-level text and are therefore only pseudo-streaming. True token-level synthesis must generate speech from uncertain prefixes while maintaining perceptual continuity over an unbounded stream with bounded context. We present X2Streaming-TTS, a causal TTS framework that consumes asynchronously arriving text tokens and emits speech without accessing future input. To handle uncertain prefixes, we introduce causal commitment, which keeps ambiguous expressions provisional through uncertainty-aware buffering and performs capacity-adaptive, punctuation-aware segmentation. To preserve acoustic continuity, we further introduce causal speech-state inheritance, which carries the complete Code2Wav state and selected historical Talker states across segment boundaries. Together with an attention prior constraint, it blocks access to future positions while retaining bounded acoustic context. Experiments show that X2Streaming-TTS outperforms existing pseudo-streaming models on most subjective and objective metrics. Further analysis shows that causal commitment stabilizes online segmentation and reduces failures caused by insufficient context, while speech-state inheritance improves boundary continuity without degrading naturalness or speaker identity. X2Streaming-TTS thus achieves strict token-level synthesis with quality comparable to the evaluated offline baselines, a median time to first audio token (TTFT) of 15.8 ms for a single request, and a median TTFT of 260.8 ms at 128 concurrent requests. Our implementation is publicly available at https://github.com/X-Square-Robot/X2Streaming-TTS .

cs.CL↗

arXiv2026

InstructVVT: Instruction-Driven Video Virtual Try-On without Auxiliary Spatial Priors

Video virtual try-on is a highly constrained editing task requiring the precise replacement of a target person's clothing while strictly preserving the original video's spatial structure and temporal dynamics. Existing methods heavily rely on auxiliary handcrafted spatial priors (e.g., masks, poses) for editing control. However, these priors are prone to failure in unconstrained real-world videos and often compress rich visual context into incomplete structural signals. Furthermore, standard reconstruction objectives fail to fully capture try-on-specific human preferences. To address these challenges, we propose InstructVVT, an instruction-driven and reference-guided video virtual try-on framework based on a Diffusion Transformer (DiT) that operates without inference-time spatial priors. Our core insight is to recover fine-grained control directly from the input triplet (source video, reference garment, and instruction) via a dual-level reference conditioning scheme. Specifically, an MLLM infers semantic edit tokens for target disambiguation and structural preservation, while a lightweight conditioning pathway explicitly injects fine-grained visual garment details. Finally, we design a try-on-specific reward and utilize the DiffusionNFT algorithm to align the model with human preferences. Extensive experiments on ViViD-S and TripVVT-Bench demonstrate that InstructVVT outperforms state-of-the-art open-source methods in garment fidelity, structural preservation, and temporal consistency, despite requiring fewer inference-time controls.

cs.CV↗

arXiv2026

KOOPMAN-Luenberger Observer Design for Nonlinear Systems with Application to the Monitoring of a Latent Thermal Energy Storage

State estimation for nonlinear dynamical systems remains a fundamental challenge, particularly when measurements are sparse and internal states are inaccessible. This work presents a KOOPMAN-based Linear State Observer (KOOPMAN-LSO) design framework that enables linear observer synthesis for nonlinear systems through KOOPMAN operator theory. The nonlinear dynamics are lifted into a higher-dimensional observable space using physics-informed basis functions, where a linear predictor with control is identified via extended dynamic mode decomposition with control (eDMDc). A discrete-time Luenberger observer is then constructed in the lifted space, and the observer gain is obtained through a dual linear - quadratic regulator (LQR) formulation to ensure stable and tunable estimation error dynamics. The proposed framework combines the representational capability of KOOPMAN lifting with the simplicity and computational efficiency of linear observer design, providing a systematic approach for nonlinear state estimation under limited sensing. Its effectiveness is demonstrated on a latent thermal energy storage (LTES) system based on phase-change materials (PCM), where internal temperature states are not directly measurable. Experimental results under varying operating conditions show accurate reconstruction of unmeasured states from limited output measurements, illustrating the potential of KOOPMAN-LSO design for practical nonlinear systems. The proposed approach achieves high-fidelity reconstruction with an RMSE as low as 0.0819 {\deg}C for the LTES outlet temperature and generally below 1.0 {\deg}C for observable internal PCM temperatures.

eess.SY↗

arXiv2026

X2-Turn: Frame-Synchronous Dual-Head Modeling for Joint Streaming ASR and Turn State Prediction

Accurate and responsive turn-taking is essential for spoken dialogue systems, which must distinguish in real time between user interruptions, backchannels that should be ignored, and the completion of an utterance. Prior modular approaches typically optimize turn state prediction at the utterance or fixed-chunk level, creating a mismatch with the continuous turn state estimate, and often depend on an auxiliary ASR model, which limits responsiveness and increases overall system complexity. Therefore, we present X2-Turn, a frame-synchronous turn state prediction method via delayed-stream modeling. Specifically, building on the pretrained Voxtral Realtime model, we introduce a frame-synchronous turn state head that operates in parallel with the ASR head on shared streaming representations, jointly predicting ASR tokens and fine-grained turn states at the frame level. Experiments on bilingual EasyTurn and Full-Duplex-Bench demonstrate that the proposed method achieves an effective trade-off between turn state accuracy and decision latency.

cs.CL↗

arXiv2026

Inevitable shock formation for 3-D compressible Euler flows

We prove that solutions arising from smooth, sufficiently small, compactly supported perturbations of non-vacuum constant states in three-dimensional irrotational compressible flow must blow up in finite time, without any symmetry assumptions or other restrictions on the initial data. Moreover, we prove that shock inevitably forms at the boundary of the maximal Cauchy development, and that its formation time agrees, in the small-data asymptotic regime, with the lifespan predicted by the radiation-field analysis.

math.AP↗

arXiv2026

SynVAR: Synergizing Spatial and Semantic Alignment in Visual Autoregressive Model

VAR has gained widespread popularity due to its next-scale prediction paradigm. However, it faces substantial performance bottlenecks when handling complex scenes with multiple objects and attributes. Existing diffusion-based enhancement methods fail to adequately address the unique challenge of cross-scale error propagation and accumulation in VAR. To this end, we propose SynVAR, the first training-free enhancement framework specifically tailored for the VAR paradigm, which introduces a spatial-semantic collaborative control strategy to effectively suppress propagation error and improve generation quality. SynVAR comprises three key components: (1) Global guidance to ensure reasonable spatial structure in the early stages, (2) Receptive field constraints to mitigate early-stage semantic confusion, (3) High-frequency compensation to recover fine-grained details. Extensive quantitative and qualitative experiments demonstrate the significant improvements in the ability of SynVAR to enhance the VAR's capability for complex scene modeling.

cs.CV↗

arXiv2026

Physics-Grounded Materials Artificial Intelligence for Reliable Materials Discovery

Artificial intelligence (AI) is transforming materials discovery, yet conventional data-driven approaches often suffer from limited interpretability, poor extrapolation, and inconsistency with physical laws. Since materials behavior is fundamentally governed by thermodynamics, kinetics, electronic structure, transport processes, and operating environments, the next generation of materials intelligence must move beyond correlation-based prediction toward physics-grounded reasoning. In this Perspective, we systematically discuss Physics-Grounded Materials AI (PhysMat AI) as a unifying perspective for integrating physical knowledge into materials intelligence through five complementary roles: physics as prior knowledge, descriptors, constraints, verifiers, and infrastructure. Using representative examples from catalysis, solid-state electrolytes in solid-state battery, and hydrogen-storage materials, we illustrate how physical principles guide data representation, model reasoning, validation workflows, and knowledge management. We further present how AI agents can leverage these physics-aware components to perform mechanism-guided discovery within physically feasible search spaces. Finally, we outline a developmental roadmap from physics-aware AI to physics-reasoning AI and ultimately physics-autonomous AI. Looking forward, materials intelligence should evolve from predictive models toward autonomous scientific systems capable of integrating physical reasoning, multiscale simulations, experimental validation, and continuous knowledge updating for reliable materials discovery.

physics.chem-ph↗

arXiv2026

MMAligner: Safeguarding Multimodal Large Language Models through Representation Calibration

Multimodal large language models (MLLMs) often refuse unsafe text prompts yet generate harmful responses to semantically equivalent multimodal inputs. Existing defenses either rely on external guardrails, which add inference overhead without repairing intrinsic flaws, or safety fine-tuning, which treats alignment as black-box optimization and may sacrifice utility or require large multimodal datasets. To identify the cause of this safety disparity, we analyze MLLM representations geometrically. We find that safety mechanisms learned from text persist across modalities: a shared safety subspace and refusal boundary remain effective, and representations inside this boundary consistently trigger refusals. However, unsafe multimodal inputs undergo a representation shift that places most of them outside the boundary, allowing them to bypass the model's intrinsic safety mechanism. This indicates that multimodal safety degradation stems from representation misalignment rather than the absence of safety capability. Based on this finding, we propose MMAligner, a safeguarding method that calibrates unsafe multimodal representations into the pre-existing refusal region. MMAligner applies a hard lower bound to ensure refusal, a soft upper bound to avoid excessive modification, and a preservation objective for benign inputs. Experiments across multiple open-source MLLMs show that MMAligner raises the average refusal rate on unsafe multimodal inputs to 99% with less than 2% utility degradation and minimal training data, substantially improving the safety-utility trade-off over existing baselines. (*Due to the notification from arXiv, "The Abstract field cannot be longer than 1,920 characters", the Abstract that appeared is shortened.)

cs.CR↗

arXiv2026

Why Ammoniated Lithium Borohydrides Liquefy and Resolidify?

Ammonia ($\mathrm{NH_3}$) absorption drives $\mathrm{LiBH_4\!\cdot\!xNH_3}$ through a re-entrant ``solid--liquid--solid'' transition: $\mathrm{LiBH_4\!\cdot\!NH_3}$ is a well-defined solid ammoniate, compositions near $\mathrm{LiBH_4\!\cdot\!2NH_3}$ are liquid-like or partially liquefied, whereas $\mathrm{LiBH_4\!\cdot\!3NH_3}$ returns to a more rigid non-liquid ammoniate state. However, the microscopic origin of this unintuitive response remains a long-lasting mystery. Here, we uncover its mechanism. Cross-database analysis identifies borohydrides as a particularly state-diverse and composition-responsive material family. Structure prediction and ab initio molecular simulations reveal that $\mathrm{NH_3}$ progressively replaces $\mathrm{BH_4^-}$ in the Li coordination shell. The liquid-like state emerges not at the highest $\mathrm{NH_3}$ loading but near $x\approx2$, where Li--N and Li--B coordination modes are strongly mixed, coordination memory is weakest, and the sampled Li--N/N$\cdots$B coordination landscape is broadest. Further ammoniation produces Li--N-dominant coordination and slows $\mathrm{BH_4^-}/\mathrm{NH_3}$ contact renewal, with the resulting increase in network persistence and accompanying recovery of a rigid ammoniate state. Pressure--composition isotherm, $^{1}\mathrm{H}$ and $^{11}\mathrm{B}$ nuclear magnetic resonance, and Raman measurements support this non-monotonic state evolution and associated $\mathrm{BH_4^-}/\mathrm{NH_3}$ reorganization. These findings transform ammonia-induced liquefaction from an empirical phase anomaly into a competition between native-network disruption, mixed-coordination frustration, and ligand-built network reconstruction, providing a framework for chemically switching between transport-favouring fluidity and stability-favouring rigidity in hydrogen-rich materials.

cond-mat.mtrl-sci↗