arXiv subjects

Yi Huang

Publications and source records attributed to Yi Huang.

At least 55 records · Page 3Linked to original sources

arXiv2026

WPT: World-to-Policy Transfer via Online World Model Distillation

Recent years have witnessed remarkable progress in world models, which primarily aim to capture the spatio-temporal correlations between an agent's actions and the evolving environment. However, existing approaches often suffer from tight runtime coupling or depend on offline reward signals, resulting in substantial inference overhead or hindering end-to-end optimization. To overcome these limitations, we introduce WPT, a World-to-Policy Transfer training paradigm that enables online distillation under the guidance of an end-to-end world model. Specifically, we develop a trainable reward model that infuses world knowledge into a teacher policy by aligning candidate trajectories with the future dynamics predicted by the world model. Subsequently, we propose policy distillation and world reward distillation to transfer the teacher's reasoning ability into a lightweight student policy, enhancing planning performance while preserving real-time deployability. Extensive experiments on both open-loop and closed-loop benchmarks show that our WPT achieves state-of-the-art performance with a simple policy architecture: it attains a 0.11 collision rate (open-loop) and achieves a 79.23 driving score (closed-loop) surpassing both world-model-based and imitation-learning methods in accuracy and safety. Moreover, the student sustains up to 4.9x faster inference, while retaining most of the gains.

cs.CV↗

arXiv2026

Prospective constraints on dark energy from nanohertz individual gravitational wave sources

Nanohertz gravitational waves (GWs) from supermassive binary black holes (SMBBHs), detectable via pulsar timing arrays (PTAs), offer a novel avenue to constrain dark energy. Based on cosmological simulations and semi-analytic galaxy formation models, this study explores the detectability of individual nanohertz SMBBH sources using next-generation PTAs and their potential for constraining dark energy under an optimistic scenario considering only the presence of white noise. By constructing light-cone SMBBH populations across hardening timescales ($τ_H = 0.1/5/10$Gyr) and computing signal-to-noise ratios (SNR), we find advanced PTAs can resolve $10^2$--$10^3$ sources with SNR $> 8$ (primarily at $z < 1$ with chirp masses of $10^8$--$10^{10}M_{\odot}$). If electromagnetic counterparts can be identified, optimal configurations ($σ_t = 50$ns, $N_p = 1000$, $T_{\text{obs}} = 30$yr with$ τ_H \leq 5$Gyr) could constrain the dark energy equation-of-state (EoS) parameter $w$ to $Δw \sim 0.023$--$0.048$, where the constraints only exhibit weak dependence on $τ_H$ within $0.1$--$5$Gyr. If only $10\%$ of GW sources have detectable electromagnetic counterparts, constraints weaken to $Δw = 0.075$ ($τ_H = 0.1$Gyr) and $Δw = 0.162$ ($τ_H = 5$Gyr) under the most optimal parameter configuration. What's more, conservative PTAs ($N_p = 500$, $σ_t = 100$--$200$ns) with additional $30$-year data accumulation could double resolvable source counts and improve $Δw$ precision by $\sim 40\%$.

astro-ph.CO↗

arXiv2026

VideoChat-M1: Collaborative Policy Planning for Video Understanding via Multi-Agent Reinforcement Learning

By leveraging tool-augmented Multimodal Large Language Models (MLLMs), multi-agent frameworks are driving progress in video understanding. However, most of them adopt static and non-learnable tool invocation mechanisms, which limit the discovery of diverse clues essential for robust perception and reasoning regarding temporally or spatially complex videos. To address this challenge, we propose a novel Multi-agent system for video understanding, namely VideoChat-M1. Instead of using a single or fixed policy, VideoChat-M1 adopts a distinct Collaborative Policy Planning (CPP) paradigm with multiple policy agents, which comprises three key processes. (1) Policy Generation: Each agent generates its unique tool invocation policy tailored to the user's query; (2) Policy Execution: Each agent sequentially invokes relevant tools to execute its policy and explore the video content; (3) Policy Communication: During the intermediate stages of policy execution, agents interact with one another to update their respective policies. Through this collaborative framework, all agents work in tandem, dynamically refining their preferred policies based on contextual insights from peers to effectively respond to the user's query. Moreover, we equip our CPP paradigm with a concise Multi-Agent Reinforcement Learning (MARL) method. Consequently, the team of policy agents can be jointly optimized to enhance VideoChat-M1's performance, guided by both the final answer reward and intermediate collaborative process feedback. Extensive experiments demonstrate that VideoChat-M1 achieves SOTA performance across eight benchmarks spanning four tasks. Notably, on LongVideoBench, our method outperforms the SOTA model Gemini 2.5 pro by 3.6% and GPT-4o by 15.6%.

cs.CV↗

arXiv2026

TARDis: Time Attenuated Representation Disentanglement for Incomplete Multi-Modal Tumor Segmentation and Classification

The accurate diagnosis and segmentation of tumors in contrast-enhanced Computed Tomography (CT) are fundamentally driven by the distinctive hemodynamic profiles of contrast agents over time. However, in real-world clinical practice, complete temporal dynamics are often hard to capture by strict radiation dose limits and inconsistent acquisition protocols across institutions, leading to a prevalent missing modality problem. Existing deep learning approaches typically treat missing phases as absent independent channels, ignoring the inherent temporal continuity of hemodynamics. In this work, we propose Time Attenuated Representation Disentanglement (TARDis), a novel physics-aware framework that redefines missing modalities as missing sample points on a continuous Time-Attenuation Curve. We first hypothesize that the latent feature can be disentangled into a time-invariant static component (anatomy) and a time-dependent dynamic component (perfusion). We achieve this via a dual-path architecture: a quantization-based path using a learnable embedding dictionary to extract consistent anatomical structures, and a probabilistic path using a Hemodynamic Conditional Variational Autoencoder to model dynamic enhancement conditioned on the estimated scan time. This design allows the network to infer missing hemodynamic features by sampling from the learned latent distribution. Extensive experiments on a large-scale multi-modal private abdominal CT dataset (2,282 patients) and two public datasets demonstrate that TARDis significantly outperforms state-of-the-art incomplete modality frameworks. Notably, our method maintains robust diagnostic performance even in extreme data-sparsity scenarios, highlighting its potential for reducing radiation exposure while maintaining diagnostic precision.

cs.CV↗

arXiv2026

Fast Physics-Driven Untrained Network for Highly Nonlinear Inverse Scattering Problems

Untrained neural networks (UNNs) offer high-fidelity electromagnetic inverse scattering reconstruction but are computationally limited by high-dimensional spatial-domain optimization. We propose a Real-Time Physics-Driven Fourier-Spectral (PDF) solver that achieves sub-second reconstruction through spectral-domain dimensionality reduction. By expanding induced currents using a truncated Fourier basis, the optimization is confined to a compact low-frequency parameter space supported by scattering measurements. The solver integrates a contraction integral equation (CIE) to mitigate high-contrast nonlinearity and a contrast-compensated operator (CCO) to correct spectral-induced attenuation. Furthermore, a bridge-suppressing loss is formulated to enhance boundary sharpness between adjacent scatterers. Numerical and experimental results demonstrate a 100-fold speedup over state-of-the-art UNNs with robust performance under noise and antenna uncertainties, enabling real-time microwave imaging applications.

cs.LG↗

arXiv2026

Unifying Deductive and Abductive Reasoning in Knowledge Graphs with Masked Diffusion Model

Deductive and abductive reasoning are two critical paradigms for analyzing knowledge graphs, enabling applications from financial query answering to scientific discovery. Deductive reasoning on knowledge graphs usually involves retrieving entities that satisfy a complex logical query, while abductive reasoning generates plausible logical hypotheses from observations. Despite their clear synergistic potential, where deduction can validate hypotheses and abduction can uncover deeper logical patterns, existing methods address them in isolation. To bridge this gap, we propose DARK, a unified framework for Deductive and Abductive Reasoning in Knowledge graphs. As a masked diffusion model capable of capturing the bidirectional relationship between queries and conclusions, DARK has two key innovations. First, to better leverage deduction for hypothesis refinement during abductive reasoning, we introduce a self-reflective denoising process that iteratively generates and validates candidate hypotheses against the observed conclusion. Second, to discover richer logical associations, we propose a logic-exploration reinforcement learning approach that simultaneously masks queries and conclusions, enabling the model to explore novel reasoning compositions. Extensive experiments on multiple benchmark knowledge graphs show that DARK achieves state-of-the-art performance on both deductive and abductive reasoning tasks, demonstrating the significant benefits of our unified approach.

cs.AI↗

arXiv2026

DRL-Enabled Trajectory Planing for UAV-Assisted VLC: Optimal Altitude and Reward Design

Recently, the integration of unmanned aerial vehicle (UAV) and visible light communication (VLC) technologies has emerged as a promising solution to offer flexible communication and efficient lighting. This letter investigates the three-dimensional trajectory planning in a UAV-assisted VLC system, where a UAV is dispatched to collect data from ground users (GUs). The core objective is to develop a trajectory planning framework that minimizes UAV flight distance, which is equivalent to maximizing the data collection efficiency. This issue is formulated as a challenging mixed-integer non-convex optimization problem. To tackle it, we first derive a closed-form optimal flight altitude under specific VLC channel gain threshold. Subsequently, we optimize the UAV horizontal trajectory by integrating a novel pheromone-driven reward mechanism with the twin delayed deep deterministic policy gradient algorithm, which enables adaptive UAV motion strategy in complex environments. Simulation results validate that the derived optimal altitude effectively reduces the flight distance by up to 35% compared to baseline methods. Additionally, the proposed reward mechanism significantly shortens the convergence steps by approximately 50%, demonstrating notable efficiency gains in the context of UAV-assisted VLC data collection.

cs.LG↗

arXiv2026

Perturbative renormalization group approach to magic-angle twisted bilayer graphene using topological heavy fermion model

We develop a perturbative renormalization group (RG) theory for the topological heavy fermion (THF) model, describing magic-angle twisted bilayer graphene (MATBG) as an emergent Anderson lattice. Our theory focuses on an energy window where the interactions can be treated perturbatively within the THF model, providing insights into the low-energy physics. In particular, the realistic parameters place MATBG near an intermediate regime where the Hubbard interaction $U$ and the hybridization energy $γ$ are comparable, motivating the need for RG analysis. Our approach analytically tracks the flow of single-particle parameters and Coulomb interactions within an energy window below $0.1$ eV, providing implications for distinguishing between Kondo-like ($U\gg γ$) and projected-limit/Mott-semimetal ($U\ll γ$) scenarios at low energies. We show that the RG flows generically lower the ratio $U/γ$ and drive MATBG toward the chiral limit, consistent with the previous numerical study based on the Bistritzer-MacDonald model. The framework presented here also applies to other moiré systems and stoichiometric materials that admit a THF description, including magic-angle twisted trilayer graphene, twisted checkerboard model, and Lieb lattice, among others, providing a foundation for developing low-energy effective theories relevant to a broad class of topological flat-band materials.

cond-mat.str-el↗

arXiv2026

Disorder-induced strong-field strong-localization in 2D systems

A recent STM experiment in 2D bilayer graphene [Y.-C. Tsui, et al., Nature 628, 287 (2024)], under a strong perpendicular magnetic field, has made a direct observation of the existence of three distinct filling-factor-dependent quantum phases in the lowest Landau level: the incompressible fractional quantum Hall liquid, a crystalline compressible hexagonal Wigner crystal with long-range order and rotational symmetry-breaking, and a random localized solid phase with no spatial order. We argue that the spatially random localized phase at low filling is the recently proposed disorder-dominated strongly localized amorphous "Anderson solid" phase [A. Babber, et al., arXiv:2601.03521], which appears generically at a sample-dependent filling factor.

cond-mat.mes-hall↗

arXiv2026

MagicFight: Personalized Martial Arts Combat Video Generation

Amid the surge in generic text-to-video generation, the field of personalized human video generation has witnessed notable advancements, primarily concentrated on single-person scenarios. However, to our knowledge, the domain of two-person interactions, particularly in the context of martial arts combat, remains uncharted. We identify a significant gap: existing models for single-person dancing generation prove insufficient for capturing the subtleties and complexities of two engaged fighters, resulting in challenges such as identity confusion, anomalous limbs, and action mismatches. To address this, we introduce a pioneering new task, Personalized Martial Arts Combat Video Generation. Our approach, MagicFight, is specifically crafted to overcome these hurdles. Given this pioneering task, we face a lack of appropriate datasets. Thus, we generate a bespoke dataset using the game physics engine Unity, meticulously crafting a multitude of 3D characters, martial arts moves, and scenes designed to represent the diversity of combat. MagicFight refines and adapts existing models and strategies to generate high-fidelity two-person combat videos that maintain individual identities and ensure seamless, coherent action sequences, thereby laying the groundwork for future innovations in the realm of interactive video content creation. Website: https://MingfuYAN.github.io/MagicFight/ Dataset: https://huggingface.co/datasets/MingfuYAN/KungFu-Fiesta

cs.CV↗

arXiv2025

ClarifyMT-Bench: Benchmarking and Improving Multi-Turn Clarification for Conversational Large Language Models

Large language models (LLMs) are increasingly deployed as conversational assistants in open-domain, multi-turn settings, where users often provide incomplete or ambiguous information. However, existing LLM-focused clarification benchmarks primarily assume single-turn interactions or cooperative users, limiting their ability to evaluate clarification behavior in realistic settings. We introduce \textbf{ClarifyMT-Bench}, a benchmark for multi-turn clarification grounded in a five-dimensional ambiguity taxonomy and a set of six behaviorally diverse simulated user personas. Through a hybrid LLM-human pipeline, we construct 6,120 multi-turn dialogues capturing diverse ambiguity sources and interaction patterns. Evaluating ten representative LLMs uncovers a consistent under-clarification bias: LLMs tend to answer prematurely, and performance degrades as dialogue depth increases. To mitigate this, we propose \textbf{ClarifyAgent}, an agentic approach that decomposes clarification into perception, forecasting, tracking, and planning, substantially improving robustness across ambiguity conditions. ClarifyMT-Bench establishes a reproducible foundation for studying when LLMs should ask, when they should answer, and how to navigate ambiguity in real-world human-LLM interactions.

cs.CL↗

arXiv2025

Hardware-Algorithm Co-Design for Hyperdimensional Computing Based on Memristive System-on-Chip

Hyperdimensional computing (HDC), utilizing a parallel computing paradigm and efficient learning algorithm, is well-suited for resource-constrained artificial intelligence (AI) applications, such as in edge devices. In-memory computing (IMC) systems based on memristive devices complement this by offering energy-efficient hardware solutions. To harness the advantages of both memristive IMC hardware and HDC algorithms, we propose a hardware-algorithm co-design approach for implementing HDC on a memristive System-on-Chip (SoC). On the hardware side, we utilize the inherent randomness of memristive crossbar arrays for encoding and employ analog IMC for classification. At the algorithm level, we develop hardware-aware encoding techniques that map data features into hyperdimensional vectors, optimizing the classification process within the memristive SoC. Experimental results in hardware demonstrate 90.71% accuracy in the language classification task, highlighting the potential of our approach for achieving energy-efficient AI deployments on edge devices.

cs.ET↗

arXiv2025

Apparent inconsistency between Streda formula and Hall conductivity in reentrant integer quantum anomalous Hall effect in twisted MoTe$_2$

Recent experiments in twisted bilayer MoTe$_2$ (tMoTe$_2$) have uncovered a rich landscape of correlated phases. In this work, we investigate the reentrant integer quantum anomalous Hall (RIQAH) states reported by F. Xu, arXiv.2504.06972 which display a notable mismatch between the Hall conductivity measured via transport and that inferred from the Streda formula. We argue that this discrepancy can be explained if the RIQAH state is a quantum Hall bubble or Wigner crystal phase, analogous to similar well-established phenomena in two-dimensional (2D) GaAs quantum wells. While this explains the RIQAH state at filling $ν= -0.63$, F. Xu et al. report that the other RIQAH state at $ν= -0.7$ has a smaller slope, necessitating a different interpretation. We propose and substantiate with analysis of the experimental data that this discrepancy arises due to a nearby resistive peak masking the true slope. Furthermore, we identify this resistive peak as a signature of a phase transition near $ν= -0.75$, possibly driven by a van Hove singularity. The anomalous Hall response and Landau fan evolution across this transition suggest a change in Fermi-surface topology and a metallic phase with a non quantized Hall response. These observations offer new insights into the nature of the RIQAH states and raise the possibility that the nearby superconducting phase may have a valley-imbalanced metal parent state.

cond-mat.str-el↗

arXiv2025

Is the Information Bottleneck Robust Enough? Towards Label-Noise Resistant Information Bottleneck Learning

The Information Bottleneck (IB) principle facilitates effective representation learning by preserving label-relevant information while compressing irrelevant information. However, its strong reliance on accurate labels makes it inherently vulnerable to label noise, prevalent in real-world scenarios, resulting in significant performance degradation and overfitting. To address this issue, we propose LaT-IB, a novel Label-Noise ResistanT Information Bottleneck method which introduces a "Minimal-Sufficient-Clean" (MSC) criterion. Instantiated as a mutual information regularizer to retain task-relevant information while discarding noise, MSC addresses standard IB's vulnerability to noisy label supervision. To achieve this, LaT-IB employs a noise-aware latent disentanglement that decomposes the latent representation into components aligned with to the clean label space and the noise space. Theoretically, we first derive mutual information bounds for each component of our objective including prediction, compression, and disentanglement, and moreover prove that optimizing it encourages representations invariant to input noise and separates clean and noisy label information. Furthermore, we design a three-phase training framework: Warmup, Knowledge Injection and Robust Training, to progressively guide the model toward noise-resistant representations. Extensive experiments demonstrate that LaT-IB achieves superior robustness and efficiency under label noise, significantly enhancing robustness and applicability in real-world scenarios with label noise.

cs.LG↗

arXiv2025

IE2Video: Adapting Pretrained Diffusion Models for Event-Based Video Reconstruction

Continuous video monitoring in surveillance, robotics, and wearable systems faces a fundamental power constraint: conventional RGB cameras consume substantial energy through fixed-rate capture. Event cameras offer sparse, motion-driven sensing with low power consumption, but produce asynchronous event streams rather than RGB video. We propose a hybrid capture paradigm that records sparse RGB keyframes alongside continuous event streams, then reconstructs full RGB video offline -- reducing capture power consumption while maintaining standard video output for downstream applications. We introduce the Image and Event to Video (IE2Video) task: reconstructing RGB video sequences from a single initial frame and subsequent event camera data. We investigate two architectural strategies: adapting an autoregressive model (HyperE2VID) for RGB generation, and injecting event representations into a pretrained text-to-video diffusion model (LTX) via learned encoders and low-rank adaptation. Our experiments demonstrate that the diffusion-based approach achieves 33\% better perceptual quality than the autoregressive baseline (0.283 vs 0.422 LPIPS). We validate our approach across three event camera datasets (BS-ERGB, HS-ERGB far/close) at varying sequence lengths (32-128 frames), demonstrating robust cross-dataset generalization with strong performance on unseen capture configurations.

cs.CV↗

arXiv2025

Resonance analysis of one-dimensional acoustic media: a propagation matrix approach

This work analyzes the scattering resonances of general acoustic media in a one-dimensional setting using the propagation matrix approach. Specifically, we characterize the resonant frequencies as the zeros of an explicit trigonometric polynomial. Leveraging Nevanlinna's value distribution theory, we establish the distribution properties of the resonances and demonstrate that their imaginary parts are uniformly bounded, which contrasts with the three-dimensional case. In two classes of high-contrast regimes, we derive the asymptotics of both subwavelength and non-subwavelength resonances with respect to the contrast parameter. Furthermore, by applying the Newton polygon method, we recover the discrete capacitance matrix approximation for subwavelength Minnaert resonances in both Hermitian and non-Hermitian cases, thereby establishing its connection to the propagation matrix framework.

math.AP↗

arXiv2025

Leveraging Contrast Information for Efficient Document Shadow Removal

Document shadows are a major obstacle in the digitization process. Due to the dense information in text and patterns covered by shadows, document shadow removal requires specialized methods. Existing document shadow removal methods, although showing some progress, still rely on additional information such as shadow masks or lack generalization and effectiveness across different shadow scenarios. This often results in incomplete shadow removal or loss of original document content and tones. Moreover, these methods tend to underutilize the information present in the original shadowed document image. In this paper, we refocus our approach on the document images themselves, which inherently contain rich information.We propose an end-to-end document shadow removal method guided by contrast representation, following a coarse-to-fine refinement approach. By extracting document contrast information, we can effectively and quickly locate shadow shapes and positions without the need for additional masks. This information is then integrated into the refined shadow removal process, providing better guidance for network-based removal and feature fusion. Extensive qualitative and quantitative experiments show that our method achieves state-of-the-art performance.

cs.CV↗

arXiv2025

Dynamical Implicit Neural Representations

Implicit Neural Representations (INRs) provide a powerful continuous framework for modeling complex visual and geometric signals, but spectral bias remains a fundamental challenge, limiting their ability to capture high-frequency details. Orthogonal to existing remedy strategies, we introduce Dynamical Implicit Neural Representations (DINR), a new INR modeling framework that treats feature evolution as a continuous-time dynamical system rather than a discrete stack of layers. This dynamical formulation mitigates spectral bias by enabling richer, more adaptive frequency representations through continuous feature evolution. Theoretical analysis based on Rademacher complexity and the Neural Tangent Kernel demonstrates that DINR enhances expressivity and improves training dynamics. Moreover, regularizing the complexity of the underlying dynamics provides a principled way to balance expressivity and generalization. Extensive experiments on image representation, field reconstruction, and data compression confirm that DINR delivers more stable convergence, higher signal fidelity, and stronger generalization than conventional static INRs.

cs.LG↗