arXiv subjects

Fei Liu

Publications and source records attributed to Fei Liu.

At least 19 recordsLinked to original sources

arXiv2026

Neural Music Enhancement with Dual Time-Frequency Spectral Representations for Prediction and Discrimination

Non-professional music recordings shared online often suffer from background noise and reverberation, degrading perceived quality and limiting reuse. This paper proposes DSME, a music enhancement model based on dual time-frequency spectral representations. Within a generative adversarial framework, DSME uses short-time Fourier transform (STFT) spectra for generation and constant-Q transform (CQT) spectra for discrimination. Leveraging STFT's fixed window, invertibility, and predictability, the generator estimates clean amplitude-phase spectra from degraded inputs and reconstructs waveforms via inverse STFT. Exploiting CQT's log-frequency, variable-window structure aligned with musical octaves, we design an octave-segmented CQT discriminator. We also introduce a chroma-spectrum loss to emphasize pitch and harmonic consistency. Experiments show DSME outperforms baselines in objective and subjective tests, validating the effectiveness of the dual-spectrum approach.

cs.SD↗

arXiv2026

PACodec: A Low-bitrate Neural Speech Codec with Parallel Additive Vector Quantization

This paper proposes PACodec, a novel low-bitrate neural speech codec based on parallel additive vector quantization (PAVQ). Unlike the mainstream residual vector quantization (RVQ) used in most neural speech codecs, where vector quantizers (VQs) are sequentially dependent, the PAVQ strategy adopted in PACodec aggregates parallel quantization results to optimize bitrate usage. Specifically, the PAVQ adopts a "global-local-global" (GLG) design: the global encoded features are quantized in parallel by multiple independent VQs, each attending to a local component of the representation, and their outputs are aggregated through addition to yield the final global quantization result for decoding. Experimental results show that PACodec, as each VQ focuses only on local information, supports smaller codebooks and reduces bitrate by 30% compared with baselines at the same decoding quality, with only minor model complexity. Further analysis shows that, owing to the GLG framework of PAVQ, the proposed PACodec is disentanglement-friendly, and each independent VQ captures different aspects of speech, e.g., content, timbre, and acoustic details, suggesting potential for application to downstream tasks such as voice conversion.

cs.SD↗

arXiv2026

AREAs-Lab: An Interactive Environment for AI-driven Requirement Elicitation for AI Systems

Building effective AI systems increasingly depends on writing high-quality task requirements, yet users often struggle to articulate the constraints, preferences, and edge cases that determine success. This problem is especially acute in AI development, where behavior is shaped not only by human expectations but also by data characteristics. We present AREAs-Lab, an interactive environment for AI-driven Requirement Elicitation for AI systems. In AREAs-Lab, an assistant iteratively refines an initially incomplete requirement by analyzing the underlying dataset and asking targeted clarification questions to uncover the user's latent intent. To study this setting systematically, we construct a synthetic benchmark grounded in 16 public datasets spanning diverse domains and task types. Each benchmark instance includes a user profile, a complete reference requirement, and an intentionally underspecified version that serves as the assistant's starting point. We further introduce an automated evaluation pipeline based on an AI-simulated user that reveals hidden information only when appropriately prompted, enabling scalable and reproducible assessment of interactive elicitation quality. AREAs-Lab provides a controlled testbed for studying how AI assistants can transform vague user goals into actionable requirements for AI systems.

cs.HC↗

arXiv2026

Beyond Average Performance: Dynamic Instance Clustering and Specialized Algorithm Design in LLM-Assisted Evolutionary Search

Large Language Model-assisted Evolutionary Search (LES) has emerged as a powerful paradigm for automated algorithm design. However, existing LES methods primarily optimize for average performance, inherently directing search effort toward instances that contribute most to this metric while leaving others poorly served, resulting in weak tail robustness and limited real-world reliability. To address this limitation, we propose Dynamic Instance Clustering and Specialized Algorithm Design (DyCA), an LES framework with a feature-free, structure-aware mechanism for constructing reliable algorithm portfolios under heterogeneous instance distributions. DyCA treats instance clustering as a co-evolving component within the search process, reusing accumulated evaluation data as feature-free signals to progressively partition instances with similar algorithmic response patterns. The uncovered clusters decompose the mixed objective into a set of structure-aware sub-objectives, thereby enabling finer-grained and more adaptive guidance for specialized algorithm design. Experimental results across four algorithm design tasks with heterogeneous instances demonstrate that DyCA outperforms state-of-the-art LES baselines, improving tail robustness by an average of 15.2\% and overall performance by 7.1\% while maintaining competitive head performance.

cs.AI↗

arXiv2026

Symmetry-Selective Strain Control of Anisotropic Magnetic Response in a Silicon FinFET Double Quantum Dot

Strain naturally develops in three-dimensional quantum-dot structures such as silicon FinFETs during fabrication and cooling. Such strain becomes especially important in a double quantum dot, because the two dots can experience different local strain and therefore acquire different magnetic responses. To understand how this dot-to-dot strain difference affects coupled hole spins, we theoretically study the local $g$ tensors of a silicon FinFET double quantum dot by combining a three-dimensional Poisson--Schr\"odinger calculation based on a six-band $k\!\cdot\!p$ model with configuration interaction. We find that the effect of strain depends on both its tensor component and its spatial symmetry. For the diagonal components $\epsilon_{yy}$ and $\epsilon_{zz}$, strain mainly changes the principal $g$ values, with only a small opening of the maximum-response axes. In contrast, the shear component $\epsilon_{yz}$ can also change the orientation of the local magnetic response. When the strain profile preserves the transverse mirror symmetry, the shear-induced rotation is strongly suppressed. Breaking this local constraint permits a pronounced off-diagonal response and rotates the principal magnetic axes. The same component- and symmetry-selected trends appear in a Zeeman-only calculation, showing that the valence-band Zeeman coupling is sufficient to generate them, while the full Hamiltonian determines their quantitative expression. Together, these results show how the tensor component and spatial symmetry of strain can be used to control both the magnitude and orientation of the magnetic response in coupled hole-spin qubits.

cond-mat.mes-hall↗

arXiv2026

ReflectVLN: Training Vision-Language Navigation Agents with Reflective Reasoning

Existing vision-language navigation methods often couple a VLM with waypoint decoders to produce multi-step action plans, but they typically lack an explicit closed-loop mechanism for tracking semantic progress, diagnosing execution failures, and recovering from error accumulation in long-horizon navigation. To address this gap, we propose ReflectVLN, an agentic VLN framework that organizes decision-making through bidirectionally interactive intention and execution agents. The intention agent performs subtask decomposition and reflection, generating executable subtask descriptions as corrective plans. Conditioned on these descriptions, the execution agent grounds them into short-horizon actions under current observations while monitoring sub-goal progress and detecting off-track behavior. Crucially, ReflectVLN enables closed-loop bidirectional communication: the execution agent emits progress and deviation signals to trigger reflection and subtask updates on demand, and the intention agent returns structured guidance that reconditions subsequent actions for recovery. To encourage temporally coherent decisions with interpretable intermediate rationales, we introduce Action Chain-of-Thought (Action-CoT), a path-conditioned dual-query training scheme for action generation. Experiments on standard VLN benchmarks show that ReflectVLN improves success rates and path efficiency under a constrained data budget, with favorable training cost and fewer high-level intention calls at inference time, while providing interpretable intermediate decisions for analysis and collaboration. Code is available at: https://github.com/AIprogrammer/ReflectVLN

cs.CV↗

arXiv2026

SVR-R1: Bootstrapping Multi-modal Reasoning with Self-verification in Reinforcement Learning

We introduce Self-Verified Reasoner (SVR-R1), a multi-turn RL framework that turns a model's own verification into a learning signal for multimodal reasoning. For each query, the model proposes an answer using the same weights, and issues a binary self-verdict (Yes/No). A 'No' triggers a second-chance rethink; a 'Yes,' or a turn cap, finalizes the output for computing the outcome-based reward. SVR-R1 is implemented with GRPO and an asynchronous multi-turn rollout framework and needs no external supervision or auxiliary critics. We evaluate SVR-R1 on vision-language reasoning benchmarks and show that it improves accuracy by a large margin over strong standard GRPO baselines. Training dynamics show decreasing reliance on verification-fewer verification turns, yet higher test accuracy-indicating that the gap between verification and generation narrows as the policy internalizes self-correction and chooses the most confident answer via our framework. SVR-R1 bridges the less explored intersection of inference-time self-refinement and RL training for VLMs, offering a simple yet effective recipe for bootstrapping multimodal reasoning. We will open-source \textbf{SVR-R1} to facilitate future research in VLMs.

cs.AI↗

arXiv2026

ABot-N1: Toward a General Visual Language Navigation Foundation Model

Visual Language Navigation foundation models aim to unify deep reasoning for grounded spatial decisions with broad versatility for diverse embodied tasks. Current approaches typically achieve this integration via monolithic policies that map observations directly to actions, yet they often suffer from coordinate drift and poor handling of long-tail semantics. Furthermore, these black-box mappings lack interpretability, hindering the simultaneous achievement of generality, robustness, and transparency. We present ABot-N1, a step toward a general Visual Language Navigation foundation model, that addresses these challenges by decoupling cognition from control via a slow-fast architecture guided by dual visual-language signals. More specifically, a slow vision-language reasoner performs explicit Chain-of-Thought reasoning while producing a pixel goal. This compact set of image-space anchor points serves as a universal interface for diverse tasks, including point-goal, object-goal, poi-goal, instruction-following, and person-following. Subsequently, a fast action expert leverages both the textual cues and the pixel guidance to generate continuous waypoints at the native control frequency. By bridging high-level intents and low-level control through pixel-grounded anchors paired with explicit linguistic traces, our approach ensures robust, generalizable, and interpretable navigation across simulation and real-world benchmarks. ABot-N1 establishes new state-of-the-art records, delivering massive gains specifically in urban-scale navigation: boosting POI arrival by 35.0% (to 77.3%) and achieving 95.4%/92.9% SR in complex indoor and outdoor scenes. It also maintains superior robustness across object-reaching, person-following, and instruction-following tasks. New Point-Goal/POI-Goal benchmarks are released as open source to advance the field of urban-scale navigation.

cs.CV↗

arXiv2026

A Pfaffian Proof and Generalization of a Conjecture of Sun Zhiwei

Let $p$ be an odd prime, let $n=(p-1)/2$, and let $\chi=(\frac{\cdot}{p})$, with $\chi(0)=0$. For $a\in\mathbb F_p^\times$ define \[ D_a(x)=\det_{1\le i,j\le n}(x+\chi(i^2-aj)), \qquad D_a^{(0)}(x)=\det_{0\le i,j\le n}(x+\chi(i^2-aj)). \] We prove \[ D_a(0)=0 \quad\Longleftrightarrow\quad p\equiv 3 \pmod 4 \quad\text{and}\quad \chi(a n!)=1. \] For $p\equiv3\pmod4$ we also give explicit Pfaffian-square factorizations of $D_a(x)$ and $D_a^{(0)}(x)$. Let $s_p=(-1)^{\lfloor(p+1)/8\rfloor}$. If $\chi(a n!)=1$, then $s_pD_a(x)/x=s_pD_a^{(0)}(x)$ is a positive integer square. If $\chi(a n!)=-1$, then there is a positive integer $\sigma$ such that \[ s_pD_a(x)=\sigma^2(nx-1),\qquad s_pD_a^{(0)}(x)=-\sigma^2\bigl(n+(2n+1)x\bigr). \] The case $a=n!$ settles Sun's Conjecture 4.1.

math.NT↗

arXiv2026

RAISE: LLM-based Automated Heuristic Design with Robust Adversary Instance Search

Automated Heuristic Design (AHD) with Large Language Models (LLMs) has shown remarkable progress in discovering high-quality heuristics. However, existing LLM-based AHD methods optimize heuristics for a fixed training instance set and may fail catastrophically when deployed under real-world distributional shifts. We propose Robust Adversary Instance Search (RAISE), a framework that integrates constrained worst-case instance search within a principled neighborhood of the training distribution into the LLM-based evolutionary search loop. RAISE treats robust AHD as a constrained adversarial instance search problem: the outer loop evolves heuristics via LLM operators, while an LLM-free inner loop efficiently identifies hard instances within an epsilon-ball around the training instance set using a basis distribution parameterization with boundary projection. Comprehensive experiments on Online Bin Packing (OBP), Online Job Shop Scheduling (OJSP), and Online Vehicle Routing (OVRP) across five distribution families demonstrate that existing LLM-based AHD methods degrade by up to 19 times under distribution shift, while RAISE consistently maintains strong performance across all tested distributions and problem scales

cs.AI↗

arXiv2026

P-K-GCN: Physics-augmented Koopman-enhanced Graph Convolutional Network for Deep Spatiotemporal Super-resolution

High-fidelity simulation of spatiotemporal dynamics is computationally prohibitive, necessitating efficient super-resolution techniques to reconstruct high-resolution data from coarse-grained inputs. Traditional data-driven methods often lack physical constraints, and simple physics-informed learning struggles with irregular spatial geometries and intricately evolving temporal dynamics. To tackle these challenges, we propose a Physics-augmented Koopman-enhanced Graph Convolutional Network (P-K-GCN) for spatiotemporal super-resolution on irregular geometries. Specifically, a continuous spline-based GCN is first designed to extract spatial dependencies directly from coarse graph, and Koopman operator theory is incorporated to project the nonlinear dynamics into a compact latent space where temporal progression is linearized. Second, we augment the optimization objective with a physics-based loss to force the data-driven reconstructions to adhere to physical laws for improving predictive fidelity and robustness. Finally, we provide a rigorous theoretical analysis, establishing that the physics augmentation and Koopman regularization mathematically guarantees a reduction in super-resolution error by diminishing Rademacher complexity and tightening generalization bounds. We evaluate our framework on reconstructing spatially high-resolution cardiac electrodynamics across a 3D heart geometry from sparse low-resolution measurements. Numerical experiments demonstrate that our method achieves superior accuracy compared to baseline models.

cs.LG↗

arXiv2026

The Role of Free-breathing GRASP MRI in Accurate Phase Matching with 4D-CT for Motion Representation in Liver Cancer Radiotherapy

Objective: To determine whether free-breathing golden-angle radial sparse parallel (GRASP) magnetic resonance imaging (MRI) can represent respiratory-induced organ motion in patients with liver malignancies undergoing stereotactic body radiation therapy (SBRT). Methods: A retrospective analysis of 54 patients undergoing liver SBRT was conducted. Four-dimensional computed tomography (4D-CT), the gold standard for motion assessment, was used to characterize liver tumor motion. Image fusion was performed between free-breathing GRASP MRI and each respiratory phase of 4D-CT using an in-house registration program, with fusion quality quantified by maximum cross-correlation coefficient (MCC). Validation involved two blinded radiation oncologists: one repeated image fusion using the Eclipse-built-in module, while the other evaluated clinical relevance on a five-point scale. Results: The 50% respiratory phase of 4D-CT achieved the highest fusion quality with GRASP MRI, showing no significant differences compared to the 30% (P = 0.106), 40% (P = 0.632), and 60% (P = 0.792) phases. In contrast, fusion quality declined significantly beyond the mid-respiratory window (30%-60%), with poor fusion at the 0%, 10%, 20%, 80%, and 90% phases (P < 0.001). Validation by radiation oncologists corroborated these findings, with the 50% phase achieving the highest score. Subjective scores remained above 4 for phases 30%-70%, while scores for the remaining phases fell below 4. Conclusion: Free-breathing GRASP MRI cannot independently represent organ motion across all respiratory phases; it accurately characterizes motion only within the mid-respiratory phases (30%-60%), with optimal performance at the 50% phase. When used as a delineation standard in liver SBRT, GRASP MRI should be combined with 4D-CT or dynamic imaging modalities to ensure comprehensive motion assessment and accurate target volume definition.

physics.med-ph↗

arXiv2026

An Ultra-Low-Bitrate Neural Speech Codec with Plain-to-Pseudo Synergistic Vector Quantization

Most neural speech codecs use residual vector quantization (RVQ), in which later VQs contribute less but consume the same bitrate, leading to inefficiency. We propose P2PSynCodec, an ultra-low-bitrate neural speech codec with a plain-to-pseudo synergistic vector quantizer (P2PSVQ). P2PSVQ consists of one plain VQ and multiple pseudo VQs. The plain VQ produces basic tokens by quantization, while the pseudo VQs generate auxiliary tokens by neural prediction and incur zero transmitted bitrate. Thus, speech is decoded from the plain-VQ tokens together with predicted pseudo-VQ tokens, greatly reducing bitrate. Experiments show that P2PSynCodec achieves speech reconstruction quality comparable to competing codecs at 2.0 kbps while operating at only 0.5 kbps, demonstrating high efficiency for ultra-low-bitrate speech coding.

eess.AS↗

arXiv2026

EpiEvolve: Self-Evolving Agents for Streaming Pandemic Forecasting under Regime Shifts

Epidemic LLM forecasters are usually trained and evaluated as static supervised models, whereas operational pandemic forecasting is a streaming process in which labels arrive after predictions and disease regimes shift over time. We study this mismatch in weekly COVID-19 hospitalization trend forecasting across five variant regimes. We introduce EpiEvolve, a self-evolving agent that wraps an LLM forecaster trained on the warm-start period and keeps its weights fixed during streaming. EpiEvolve adapts by storing forecast outcomes in a hierarchical episodic memory, reflecting on delayed labels, retrieving cases relevant to the current regime, and distilling recurring errors into strategic rules. The resulting context lets the forecaster reuse its own past predictions and outcomes in later weeks while following a chronological protocol that prevents future leakage. On the streaming dataset, EpiEvolve reaches $0.629$ average accuracy, compared with $0.561$ for the static backbone and $0.325$ for the external CDC ensemble, and reduces recovery lag after regime shifts from $5$ to $2$ weeks. Ablations show that reflection, strategic memory, and regime-aware retrieval each contribute to the gains.

cs.AI↗

arXiv2026

LLM-Enabled Automated Algorithm Design for Multiuser Fluid Antenna Communications

Fluid antenna is a new reconfigurable antenna technology that can dynamically adjust the positions or ports of radiating elements and therefore provides a new degree of freedom for wireless communications. However, the associated port selection is a challenging large-scale combinatorial optimization problem and difficult to solve. Existing manually designed heuristic algorithms are not only labor-intensive, but cannot achieve satisfactory performance. In this paper, we propose a novel paradigm that leverages large language models (LLMs) for automated design of optimization algorithms for fluid antenna systems without manual hyperheuristic tuning. Specifically, we study the problem of maximizing the minimum signal-to-interference-plus-noise ratio (SINR) in the downlink to ensure fairness among users by optimizing port selection and beamforming. We investigate two LLM-enabled algorithm optimization strategies. The first is to optimize the crossover and mutation operations to enhance the performance of the well-known genetic algorithm and the second is to design AutoPort, a new heuristic from scratch by LLM, to solve the optimization problem. Simulation results verify that the proposed method can achieve near-optimal performance and significant improvement over the conventional genetic algorithm and the deep learning approach.

cs.IT↗

arXiv2026

Large Deviation Functions for Open Quantum Systems with a Strong Symmetry

In open quantum systems with strong symmetries, the global scaled cumulant generating function (SCGF) is generally nonanalytic, so the G\"artner-Ellis theorem cannot directly yield the genuine large-deviation rate function. To address this issue, we propose that the theorem remains valid within blocks of the systems' operator space: we first obtain local rate functions for each block via the theorem and then recover the global one by minimization. This approach is justified by the dissipative freezing phenomenon in such systems. We demonstrate the scheme in an analytical model and a three-spin model with XX interaction. In the latter, we find that the vanishing of a nonanalytic point in the global SCGF under dephasing appears as an avoided ``level'' crossing, and we quantify this behavior using a degenerate perturbation theory.

cond-mat.stat-mech↗

arXiv2026

TripVVT: A Large-Scale Triplet Dataset and a Coarse-Mask Baseline for In-the-Wild Video Virtual Try-On

Due to the scarcity of large-scale in-the-wild triplet data and the improper use of masks, the performance of video virtual try-on models remains limited. In this paper, we first introduce **TripVVT-10K**, the largest and most diverse in-the-wild triplet dataset to date, providing explicit video-level cross-garment supervision that existing video datasets lack. Built upon this resource, we develop **TripVVT**, a Diffusion Transformer-based framework that replaces fragile garment masks with a simple, stable human-mask prior, enabling reliable background preservation while remaining robust to real-world motion, occlusion, and cluttered scenes. To support comprehensive evaluation, we further establish **TripVVT-Bench**, a 100-case benchmark covering diverse garments, complex environments, and multi-person scenarios, with metrics spanning video quality, try-on fidelity, background consistency, and temporal coherence. Compared to state-of-the-art academic and commercial systems, TripVVT achieves superior video quality and garment fidelity while markedly improving generalization to challenging in-the-wild videos. We publicly release the dataset and benchmark, which we believe provide a solid foundation for advancing controllable, realistic, and temporally stable video virtual try-on.

cs.CV↗

arXiv2026

Agent-Centric Observation Adaptation for Robust Visual Control under Dynamic Perturbations

Real-world visual systems face time-varying perturbations, including weather, sensor noise, compression artifacts, and background distractions. Existing image restoration methods are typically designed for fixed corruption types and optimized for pixel-level fidelity, leaving open two questions: how restoration behaves under non-stationary corruption switching, and whether pixel-level fidelity preserves the task-relevant information needed by downstream models. To study this setting, we introduce the Visual Degraded Control Suite (VDCS), a benchmark that injects Markov-switching physical degradations into rendered scenes. We further identify a fundamental failure mode of reconstruction-based representations: faithfully reconstructing corrupted observations forces the latent state to encode corruption-specific nuisance information, thereby contaminating downstream models. From an information-bottleneck perspective, anchoring the representation to the clean foreground eliminates this contamination. Motivated by this analysis, we propose \emph{Agent-Centric Observations with Mixture-of-Experts} (ACO-MoE), a frozen, plug-and-play observation adapter that combines a routed bank of restoration experts with a foreground-mask branch. ACO-MoE is pretrained entirely offline on synthetic rendered data with automatically generated degradation pairs and simulation-derived foreground masks, requiring no manual annotation. At inference time, it takes only corrupted RGB as input without corruption labels, clean reference frames, or foreground masks. Across VDCS, DMC-GB, and RoboSuite, ACO-MoE consistently improves downstream control with both model-free and model-based backbones, recovering 95.3\% of clean-input performance under challenging Markov-switching corruptions. It also generalizes zero-shot to unseen visual perturbations excluded from adapter pretraining.

cs.RO↗