Searcharxiv⌕ Search

arXiv subjects

Feng Liu

Publications and source records attributed to Feng Liu.

At least 127 records · Page 7Linked to original sources
arXiv2026

Discrete Preference Learning for Personalized Multimodal Generation

The emergence of generative models enables the creation of texts and images tailored to users' preferences. Existing personalized generative models have two critical limitations: lacking a dedicated paradigm for accurate preference modeling, and generating unimodal content despite real-world multimodal-driven user interactions. Therefore, we propose personalized multimodal generation, which captures modal-specific preferences via a dedicated preference model from multimodal interactions, and then feeds them into downstream generators for personalized multimodal content. However, this task presents two challenges: (1) Gap between continuous preferences from dedicated modeling and discrete token inputs intrinsic to generator architectures; (2) Potential inconsistency between generated images and texts. To tackle these, we present a two-stage framework called Discrete Preference learning for Personalized Multimodal Generation (DPPMG). In the first stage, to accurately learn discrete modal-specific preferences, we introduce a modal-specific graph neural network (a dedicated preference model) to learn users' modal-specific preferences, which preferences are then quantized into discrete preference tokens. In the second stage, the discrete modal-specific preference tokens are injected into downstream text and image generators. To further enhance cross-modal consistency while preserving personalization, we design a cross-modal consistent and personalized reward to fine-tune token-associated parameters. Extensive experiments on two real-world datasets demonstrate the effectiveness of our model in generating personalized and consistent multimodal content.

cs.IR↗
arXiv2026

Modeling of Far-Field Quantum Coherence by Dielectric Bodies Based on the Volume Integral Equation Method

The Hong-Ou-Mandel (HOM) effect is a hallmark of nonclassical two-photon interference. This paper develops a unified theory-numerics framework to compute angle-resolved far-field two-photon correlations from arbitrary lossless dielectric scatterers. We describe the input-output relation using a multi-channel scattering formulation that maps two populated incident channels to two selected far-field detection modes, yielding a compact two-channel transfer relation for second-order correlation function and time-domain coincidence counts. The required transfer coefficients are extracted from classical far-field complex amplitudes computed by an fast Fourier transform-accelerated volume integral equation solver, avoiding perfectly matched layers and near-to-far-field post-processing. The method is validated against analytical results for dielectric spheres and demonstrated on a polarization-converting Pancharatnam-Berry-phase metasurface, revealing strong angular dependence of quantum interference and its direct impact on HOM-dip visibility. The framework provides an efficient and physically transparent tool for structure-dependent quantum-correlation analysis, with potential applications in scatterers-enabled quantum state engineering and quantum inverse design.

quant-ph↗
arXiv2026

Neural Distribution Prior for LiDAR Out-of-Distribution Detection

LiDAR-based perception is critical for autonomous driving due to its robustness to poor lighting and visibility conditions. Yet, current models operate under the closed-set assumption and often fail to recognize unexpected out-of-distribution (OOD) objects in the open world. Existing OOD scoring functions exhibit limited performance because they ignore the pronounced class imbalance inherent in LiDAR OOD detection and assume a uniform class distribution. To address this limitation, we propose the Neural Distribution Prior (NDP), a framework that models the distributional structure of network predictions and adaptively reweights OOD scores based on alignment with a learned distribution prior. NDP dynamically captures the logit distribution patterns of training data and corrects class-dependent confidence bias through an attention-based module. We further introduce a Perlin noise-based OOD synthesis strategy that generates diverse auxiliary OOD samples from input scans, enabling robust OOD training without external datasets. Extensive experiments on the SemanticKITTI and STU benchmarks demonstrate that NDP substantially improves OOD detection performance, achieving a point-level AP of 61.31% on the STU test set, which is more than 10$\times$ higher than the previous best result. Our framework is compatible with various existing OOD scoring formulations, providing an effective solution for open-world LiDAR perception.

cs.CV↗
arXiv2026

Adapting in the Dark: Efficient and Stable Test-Time Adaptation for Black-Box Models

Test-Time Adaptation (TTA) for black-box models accessible only via APIs remains a largely unexplored challenge. Existing approaches such as post-hoc output refinement offer limited adaptive capacity, while Zeroth-Order Optimization (ZOO) enables input-space adaptation but faces high query costs and optimization challenges in the unsupervised TTA setting. We introduce BETA (Black-box Efficient Test-time Adaptation), a framework that addresses these limitations by employing a lightweight, local white-box steering model to create a tractable gradient pathway. Through a prediction harmonization technique combined with consistency regularization and prompt learning-oriented filtering, BETA enables stable adaptation with no additional API calls and negligible latency beyond standard inference. On ImageNet-C, BETA achieves a +7.1% accuracy gain on ViT-B/16 and +3.4% on CLIP, surpassing strong white-box and gray-box methods including TENT and TPT. On a commercial API, BETA achieves comparable performance to ZOO at 250x lower cost while maintaining real-time inference speed, establishing it as a practical and efficient solution for real-world black-box TTA.

cs.LG↗
arXiv2026

A Methanol-mediated Room-Temperature Synthesis of Tellurium-Loaded Liquid Scintillators for Neutrinoless Double Beta Decay Search

This study establishes a methanol-mediated room-temperature synthesis approach for tellurium-diol (Te-diol) compounds for use in tellurium-loaded liquid scintillator (Te-LS). The synthesis involves the direct reaction of telluric acid with diols (e.g., 1,2-hexanediol) in methanol (MeOH) under ambient conditions (25$\pm$5\textdegree C), with the key features of lower energy consumption and enhanced safety compared with high-temperature azeotropic distillation method. Mechanistic studies reveal that MeOH serves not merely as a solvent but also exhibits a catalytic effect, playing a dual role in this water-free, heterogeneous room-temperature synthesis. The Te-diol compounds enable fabrication of high-performance Te samples exhibiting exceptional optical transparency (attenuation length = 20.1$\pm$1.1 m at $λ$=430 nm for 1\% Te mass loading), which is reported here for the first time. Furthermore, the Te-LS achieves long-term spectral stability approaching or exceeding one year for both 1\% and 3\% Te mass loadings, and demonstrates a light yield comparable those of both the azeotropic distillation method and the SNO+ collaboration's Type I loading method, albeit modestly lower than that of their Type II method. The developed protocol offers the potential for a more energy efficient alternative for large-scale Te-LS production, particularly valuable for next-generation neutrinoless double-beta decay experiments.

physics.ins-det↗
arXiv2026

A Unified Perspective on Adversarial Membership Manipulation in Vision Models

Membership inference attacks (MIAs) aim to determine whether a specific data point was part of a model's training set, serving as effective tools for evaluating privacy leakage of vision models. However, existing MIAs implicitly assume honest query inputs, and their adversarial robustness remains unexplored. We show that MIAs for vision models expose a previously overlooked adversarial surface: adversarial membership manipulation, where imperceptible perturbations can reliably push non-member images into the "member" region of state-of-the-art MIAs. In this paper, we provide the first unified perspective on this phenomenon by analyzing its mechanism and implications. We begin by demonstrating that adversarial membership fabrication is consistently effective across diverse architectures and datasets. We then reveal a distinctive geometric signature - a characteristic gradient-norm collapse trajectory - that reliably separates fabricated from true members despite their nearly identical semantic representations. Building on this insight, we introduce a principled detection strategy grounded in gradient-geometry signals and develop a robust inference framework that substantially mitigates adversarial manipulation. Extensive experiments show that fabrication is broadly effective, while our detection and robust inference strategies significantly enhance resilience. This work establishes the first comprehensive framework for adversarial membership manipulation in vision models.

cs.CV↗
arXiv2026

User-Aware Conditional Generative Total Correlation Learning for Multi-Modal Recommendation

Multi-modal recommendation (MMR) enriches item representations by introducing item content, e.g., visual and textual descriptions, to improve upon interaction-only recommenders. The success of MMR hinges on aligning these content modalities with user preferences derived from interaction data, yet dominant practices based on disentangling modality-invariant preference-driving signals from modality-specific preference-irrelevant noises are flawed. First, they assume a one-size-fits-all relevance of item content to user preferences for all users, which contradicts the user-conditional fact of preferences. Second, they optimize pairwise contrastive losses separately toward cross-modal alignment, systematically ignoring higher-order dependencies inherent when multiple content modalities jointly influence user choices. In this paper, we introduce GTC, a conditional Generative Total Correlation learning framework. We employ an interaction-guided diffusion model to perform user-aware content feature filtering, preserving only personalized features relevant to each individual user. Furthermore, to capture complete cross-modal dependencies, we optimize a tractable lower bound of the total correlation of item representations across all modalities. Experiments on standard MMR benchmarks show GTC consistently outperforms state-of-the-art, with gains of up to 28.30% in NDCG@5. Ablation studies validate both conditional preference-driven feature filtering and total correlation optimization, confirming the ability of GTC to model user-conditional relationships in MMR tasks. The code is available at: https://github.com/jingdu-cs/GTC.

cs.IR↗
arXiv2026

Distributional Statistics Restore Training Data Auditability in One-step Distilled Diffusion Models

The proliferation of diffusion models trained on web-scale, provenance-uncertain image collections has made it essential, yet technically unresolved, to determine whether a model has learned from specific copyrighted data without authorization. Current methods primarily rely on the memorization effect, whereby models reconstruct their training images better than unseen ones, to detect unauthorized training data on a per-instance basis. This effect, however, vanishes under distillation, the now-dominant deployment pipeline that compresses compute-intensive teacher diffusion models into efficient {\em student one-step generators} mimicking the teacher's output for real-time user access. As the students train exclusively on teacher-generated outputs and never directly see the teacher's original training data, they carry no per-instance memorization of that upstream data, creating a model laundering loophole that severs the auditable link between a deployed model and its upstream training data. We nonetheless reveal that a distributional memory chain survives under distillation: the student's output distribution remains closer to the teacher's training distribution than to any non-training reference, even if no single training instance is memorized. Exploiting this chain, we develop a distributional unauthorized training data detector, grounded in kernel-based distribution discrepancy, that determines if a candidate dataset of unknown composition is statistically aligned with the student-generated distribution more than held-out non-training datasets, thus tracing provenance back to the teacher's training data. Evaluation across benchmarks and distillation setups confirms reliable detection even when unauthorized data forms a minority of the candidate set, establishing distribution-level auditing as a countermeasure to model laundering and a paradigm for accountable generative AI ecosystems.

cs.LG↗
arXiv2026

NeuroNarrator: A Generalist EEG-to-Text Foundation Model for Clinical Interpretation via Spectro-Spatial Grounding and Temporal State-Space Reasoning

Electroencephalography (EEG) provides a non-invasive window into neural dynamics at high temporal resolution and plays a pivotal role in clinical neuroscience research. Despite this potential, prevailing computational approaches to EEG analysis remain largely confined to task-specific classification objectives or coarse-grained pattern recognition, offering limited support for clinically meaningful interpretation. To address these limitations, we introduce NeuroNarrator, the first generalist EEG-to-text foundation model designed to translate electrophysiological segments into precise clinical narratives. A cornerstone of this framework is the curation of NeuroCorpus-160K, the first harmonized large-scale resource pairing over 160,000 EEG segments with structured, clinically grounded natural-language descriptions. Our architecture first aligns temporal EEG waveforms with spatial topographic maps via a rigorous contrastive objective, establishing spectro-spatially grounded representations. Building on this grounding, we condition a Large Language Model through a state-space-inspired formulation that integrates historical temporal and spectral context to support coherent clinical narrative generation. This approach establishes a principled bridge between continuous signal dynamics and discrete clinical language, enabling interpretable narrative generation that facilitates expert interpretation and supports clinical reporting workflows. Extensive evaluations across diverse benchmarks and zero-shot transfer tasks highlight NeuroNarrator's capacity to integrate temporal, spectral, and spatial dynamics, positioning it as a foundational framework for time-frequency-aware, open-ended clinical interpretation of electrophysiological data.

eess.SP↗
arXiv2026

Dispatch-Embedded Long-Term Tail Risk Assessment and Mitigation via CVaR for Renewable Power Systems

Renewable energy (RE) generation exhibits pronounced seasonality and variability, and neglecting these features can lead to significant underestimation of long-term power system risks in power supply. While long-term dispatch strategies are essential for evaluating and mitigating tail risks, they are often excluded from existing models due to their complexity. This paper proposes a long-term tail risk assessment and mitigation framework for renewable power systems, explicitly embedding dispatch strategies. A representative scenario generation method is designed, combining multi-timescale Copula modeling to capture RE's long-range variability and correlation. Building on these scenarios, an evolution-based risk assessment model is established, where Conditional Value-at-Risk (CVaR) is employed as a robust metric to quantify tail risks. Finally, a controlled evolution-based risk mitigation scheme is introduced to refine long-term dispatch strategies for mitigating tail risks. Case studies on a modified IEEE-39 bus system incorporating real-world data substantiate the efficacy of the proposed method.

eess.SY↗
arXiv2026

Prime Once, then Reprogram Locally: An Efficient Alternative to Black-Box Service Model Adaptation

Adapting closed-box service models (i.e., APIs) for target tasks typically relies on reprogramming via Zeroth-Order Optimization (ZOO). However, this standard strategy is known for extensive, costly API calls and often suffers from slow, unstable optimization. Furthermore, we observe that this paradigm faces new challenges with modern APIs (e.g., GPT-4o). These models can be less sensitive to the input perturbations ZOO relies on, thereby hindering performance gains. To address these limitations, we propose an Alternative efficient Reprogramming approach for Service models (AReS). Instead of direct, continuous closed-box optimization, AReS initiates a single-pass interaction with the service API to prime an amenable local pre-trained encoder. This priming stage trains only a lightweight layer on top of the local encoder, making it highly receptive to the subsequent glass-box (white-box) reprogramming stage performed directly on the local model. Consequently, all subsequent adaptation and inference rely solely on this local proxy, eliminating all further API costs. Experiments demonstrate AReS's effectiveness where prior ZOO-based methods struggle: on GPT-4o, AReS achieves a +27.8% gain over the zero-shot baseline, a task where ZOO-based methods provide little to no improvement. Broadly, across ten diverse datasets, AReS outperforms state-of-the-art methods (+2.5% for VLMs, +15.6% for standard VMs) while reducing API calls by over 99.99%. AReS thus provides a robust and practical solution for adapting modern closed-box models.

cs.CV↗
arXiv2026

Beyond Localization: Recoverable Headroom and Residual Frontier in Repository-Level RAG-APR

Repository-level automated program repair (APR) increasingly treats stronger localization as the main path to better repair. We ask a more targeted question: once localization is strengthened, which post-localization levers still provide recoverable gains, which are bounded within our protocol, and what residual frontier remains? We study this question on SWE-bench Lite with three representative repository-level RAG-APR paradigms, Agentless, KGCompass, and ExpeRepair. Our protocol combines Oracle Localization, within-pool Best-of-K, fixed-interface added context probes with per-condition same-token filler controls and same-repository hard negatives, and a common-wrapper oracle check. Oracle Localization improves all three systems, but Oracle success still stays below 50%. Extra candidate diversity still helps inside the sampled 10-patch pools, but that headroom saturates quickly. Under the two fixed interfaces, most informative added context conditions still outperform their own matched controls. The common-wrapper check shows different system responses: under a common wrapper, gains remain large for KGCompass and ExpeRepair, while Agentless changes more with builder choice. Prompt-level fusion still leaves a large residual frontier: the best fixed probe adds only 6 solved instances beyond the native three-system Solved@10 union. Overall, stronger localization, bounded search, evidence quality, and interface design all shape repository-level repair outcomes.

cs.SE↗
arXiv2026

Securing the Skies: A Comprehensive Survey on Anti-UAV Methods, Benchmarking, and Future Directions

Unmanned Aerial Vehicles (UAVs) are indispensable for infrastructure inspection, surveillance, and related tasks, yet they also introduce critical security challenges. This survey provides a wide-ranging examination of the anti-UAV domain, centering on three core objectives-classification, detection, and tracking-while detailing emerging methodologies such as diffusion-based data synthesis, multi-modal fusion, vision-language modeling, self-supervised learning, and reinforcement learning. We systematically evaluate state-of-the-art solutions across both single-modality and multi-sensor pipelines (spanning RGB, infrared, audio, radar, and RF) and discuss large-scale as well as adversarially oriented benchmarks. Our analysis reveals persistent gaps in real-time performance, stealth detection, and swarm-based scenarios, underscoring pressing needs for robust, adaptive anti-UAV systems. By highlighting open research directions, we aim to foster innovation and guide the development of next-generation defense strategies in an era marked by the extensive use of UAVs.

cs.CV↗
arXiv2026

A Multi-physics Alternating Coupled Inversion Using Gravity and Full Waveform Data in Salt Dome

Complex salt geometries and strong velocity contrasts pose significant challenges for velocity model building and subsalt imaging. Although full waveform inversion (FWI) provides high-resolution velocity models, its performance strongly depends on the accuracy of initial model. On the other hand, gravity focusing inversion (GFI) can recover compact density distributions and provide reliable long-wavelength structural information for seismic exploration, but it suffers from poor depth resolution and inherent non-uniqueness. To better invert salt structure by leveraging the complementary advantages of full waveform and gravity data, we propose a multi-physics alternating coupled inversion strategy for salt dome model. The proposed strategy mainly includes three parts. First, we perform FWI using a simple layered velocity model to obtain preliminary velocity updates and extract the salt top boundary. Second, this structural information is used as a constraint in GFI to recover a compact salt density distribution beneath the salt top. Third, the resulting salt geometry is used to construct an improved velocity model for the next stage of FWI. Through iterative alternation, FWI provides reliable structural constraints for GFI, while GFI supplies a more reasonable macroscopic salt model for FWI, effectively mitigating the strong dependence on the initial model. In addition, a depth-varying density contrast is introduced in GFI to better represent sediment compaction effects. Compared with unconstrained GFI and conventional FWI using a horizontally layered initial model, the proposed method effectively improves both velocity and density reconstruction in the modified BP salt model and SEG/EAGE salt model.

physics.geo-ph↗
arXiv2026

From 3D Pose to Prose: Biomechanics-Grounded Vision--Language Coaching

We present BioCoach, a biomechanics-grounded vision--language framework for fitness coaching from streaming video. BioCoach fuses visual appearance and 3D skeletal kinematics, through a novel three-stage pipeline: an exercise-specific degree-of-freedom selector that focuses analysis on salient joints; a structured biomechanical context that pairs individualized morphometrics with cycle and constraint analysis; and a vision--biomechanics conditioned feedback module that applies cross-attention to generate precise, actionable text. Using parameter-efficient training that freezes the vision and language backbones, BioCoach yields transparent, personalized reasoning rather than pattern matching. To enable learning and fair evaluation, we augment QEVD-fit-coach with biomechanics-oriented feedback to create QEVD-bio-fit-coach, and we introduce a biomechanics-aware LLM judge metric. BioCoach delivers clear gains on QEVD-bio-fit-coach across lexical and judgment metrics while maintaining temporal triggering; on the original QEVD-fit-coach, it improves text quality and correctness with near-parity timing, demonstrating that explicit kinematics and constraints are key to accurate, phase-aware coaching.

cs.CV↗
arXiv2026

Anomalous thermoelectric Hall response of interacting 2D Dirac fermions

We study the anomalous thermoelectric Hall response of two-dimensional massive Dirac fermions to first order in the electron-electron interaction. We compute both the Nernst response to a Luttinger-type gravitational potential and the particle magnetization, the latter being required to remove spurious non-transport contributions. We show that, for arbitrary interactions, the magnetization is described by a remarkably simple formula. Surprisingly, and contrary to expectations, subtracting the magnetization currents does not make the thermoelectric Hall coefficient vanish in the zero-temperature limit. We attribute this to violation of locality on the smallest length scales, which is inevitable in a quantized field theory, that happens to manifest itself in infrared physics.

cond-mat.mes-hall↗
arXiv2026

TRACE: Object Motion Editing in Videos with First-Frame Trajectory Guidance

We study object motion path editing in videos, where the goal is to alter a target object's trajectory while preserving the original scene content. Unlike prior video editing methods that primarily manipulate appearance or rely on point-track-based trajectory control, which is often challenging for users to provide during inference, especially in videos with camera motion, we offer a practical, easy-to-use approach to controllable object-centric motion editing. We present Trace, a framework that enables users to design the desired trajectory in a single anchor frame and then synthesizes a temporally consistent edited video. Our approach addresses this task with a two-stage pipeline: a cross-view motion transformation module that maps first-frame path design to frame-aligned box trajectories under camera motion, and a motion-conditioned video re-synthesis module that follows these trajectories to regenerate the object while preserving the remaining content of the input video. Experiments on diverse real-world videos show that our method produces more coherent, realistic, and controllable motion edits than recent image-to-video and video-to-video methods.

cs.CV↗
arXiv2026

Data-Driven Modal Decomposition Analysis of Unsteady Flow in a Multi-Stage Turbine

Two data-driven modal analysis approaches, proper orthogonal decomposition (POD) and dynamic mode decomposition (DMD), are applied to analyze the unsteady flow obtained by solving the Reynolds-averaged Navier-Stokes (RANS) equations in a 1.5-stage axial turbine. The reduced-order reconstructed pressure, dominant mode shapes, and dynamic features of these dominant modes in the downstream stator of the turbine are compared between POD and four DMD variants. It is found that the DMD methods based on the amplitude criterion, the Tissot criterion, and the sparsity-promoting DMD (SP-DMD) achieve reconstruction accuracy comparable to that of POD, while the frequency criterion proves unsuitable for the present problem. The second and third POD and DMD modes capture the dominant pressure fluctuation structures within the stator, and there is similarity between the corresponding POD and DMD spatial modes. The unsteady flow is primarily governed by neutral DMD modes characterized by high amplitudes and low frequencies corresponding to the basic and harmonic frequencies driven by the rotor passing frequency. While the POD analysis provides accurate reconstruction for the original snapshots, the time evolution of each POD mode does not reflect the true dynamic feature of the system. In particular, they misrepresent the fundamental frequencies of the problem. In addition, the correlations between the dominant modes in the downstream stator and the turbine adiabatic efficiency are explored across different stator clocking configurations. It is found that a clocking configuration with higher adiabatic efficiency at 50% span corresponds to a larger spatial and time component of the second and third DMD mode pair, and similarly a larger second POD mode.

physics.flu-dyn↗