Searcharxiv⌕ Search

arXiv subjects

Xinyang Li

Publications and source records attributed to Xinyang Li.

At least 19 recordsLinked to original sources
arXiv2026

TaoMate-FVSR: Audio-Guided Facial Video Restoration

Low-quality facial videos suffer from missing fine details, while AI-generated videos may additionally contain implausible teeth and lip movements that do not match the speech. Restoring the mouth region is particularly challenging: enhancing existing details can preserve structural errors, whereas regenerating them risks changing facial identity and articulation. We propose an audio-guided facial video restoration framework that reconstructs unreliable mouth details and aligns lip movements with the speech while preserving facial identity and temporal consistency. Built on a video diffusion transformer, our framework mixes input video latents with noise to retain useful source information while allowing corrupted details to be regenerated. Audio guides lip movements, reference images constrain facial appearance, and previously generated frames promote temporal consistency. We train the model in two stages, first learning visual restoration from degraded images and videos, then learning audio-guided lip correction from degraded dubbed videos. The resulting model supports both restoration with the original audio and visual dubbing with replacement audio. On talking-head restoration benchmarks, our method achieves the lowest LPIPS and DISTS and the best lip synchronization scores among the evaluated restoration baselines. Qualitative results on AI-generated videos further show improvements in facial detail and mouth-region appearance, including clearer tooth boundaries and lip contours.

cs.CV↗
arXiv2026

TaoMate-HD: Progressive-Resolution Autoregressive Audio-Video Generation

Streaming digital humans produce detailed video while continuing speech as new text arrives. High-resolution denoising is costly, and text updates can cause repeated or interrupted speech. We present TaoMate-HD, an autoregressive model for high-resolution audio-video streaming. For each short segment, or chunk, two denoising steps establish coarse video structure at low resolution. A final step refines the video to high resolution, with audio generated throughout the same trajectory. Each stage passes its audiovisual predictions to the same stage of the next chunk as history. Across text updates, the model retains this history to infer what has already been spoken. Continuation training utilizes asymmetric text conditioning, with separate text windows for generation and distillation. Experiments show a 1.6x speedup over direct high-resolution generation with the same model and hardware. Optimized stage-parallel deployment provides a further 1.9x overall speedup over serial progressive generation. Our continuation training reduces character error rate by 11.18 percentage points relative to the baseline. The model also supports hour-long audiovisual generation. Our project page is https://taoliveaigc.github.io/TaoMate-HD.

cs.CV↗
arXiv2026

Beyond Division of Labor: How Roles Shape Feedback and Recovery in Multi-Agent Systems

Roles in multi-agent large language model (LLM) systems allocate decisions as well as work. In a Planner--Executor--Reviewer team, the Reviewer's approval can determine when the team submits an answer and receives corrective feedback. We study both functions by varying role instructions while holding agents, topology, call budget, and ground-truth evaluator constant, using three open-weight models on Omni-MATH and two on SciBench. Removing roles largely eliminates functional specialization without lowering first-pass accuracy. After a failed first answer, however, submission decisions matter: a role-structured Reviewer often withholds approval, so its team submits less often and receives fewer evaluator verdicts. In an exploratory comparison with feedback available to both teams, the role-free team recovers more often across all five model--benchmark settings. Two further analyses support the importance of feedback opportunities. In a post hoc comparison at equal numbers of verdicts, the pooled recovery difference is small (role-structured minus role-free: $+0.009$, 95% interval $[-0.014,+0.033]$). In an exploratory intervention on $400$ Omni-MATH problems with gemma-4, keeping the role prompts but forcing submission in every cycle raises recovery from $0.198$ to $0.613$; the paired comparison detects no difference from the role-free team. The takeaway is to evaluate roles through recovery after failure as well as specialization. For systems with reliable external feedback, this requires measuring who may submit, how often the team receives feedback, and which agents can act on it.

cs.MA↗
arXiv2026

Wrong but Useful: Trajectory Value Beyond Answer Correctness in Multi-Agent Messages

A solution can end with a wrong answer and still help another solver. In multi-agent reasoning, evaluating only a message's proposed answer can therefore miss its contribution to the receiver. We measure this contribution as trajectory value: the change in a fixed integrator's probability of answering correctly when a cached message is present rather than removed. Controlled replay across five benchmarks and two model families separates message effects from identical-input variation. The strongest evidence is for Gemma: in a pre-registered fresh-call re-test, 34 of 73 selected wrong-answer messages pass individual confirmation, exceeding control-calibrated upper bounds on expected false confirmations of 11.4--20.8. The finding is benchmark-dependent and the conservative criterion no longer passes without SciBench. A blinded two-author audit supports the selected answer labels, while neutral replacements show that the helpful effect is not explained solely by message length and position. On held-out replay blocks of the same problems, a message-specific removal rule improves accuracy over the strongest tested correctness-only rule by 1.61 percentage points [1.01, 2.25]. Post hoc analysis localizes most of this gap to pools containing no correct proposal. Evidence for gpt-oss is weaker, and the tested predictors do not establish a practical way to identify useful errors in advance. The results support measuring answer correctness and receiver benefit separately, with value understood as conditional on the message pool, receiver and generation format.

cs.AI↗
arXiv2026

Implicit Feature Reweighting for Label-Free Graph Generalization

Graph-level out-of-distribution (OOD) generalization is critical for deploying graph neural networks in real-world applications where distribution shifts are inevitable. Existing approaches typically require class labels during representation learning to decide which substructures are predictive, limiting their applicability. We propose SCARCE, a label-free representation-learning framework that yields distribution-shift-robust graph representations from an environment partition of the training graphs alone, whether that partition comes from dataset metadata or, when none is available, from clustering a self-supervised embedding. Our method operates in two stages: (1) a variational graph autoencoder is warm-started for graph reconstruction; and (2) an adversarial confusion mechanism forces the encoder to produce representations from which a discriminator cannot distinguish environments, while preserving graph-reconstruction fidelity. We provide theoretical analysis showing that environment confusion contracts environment-discriminative directions in the representation space, which, under regularized downstream classifiers, is equivalent to imposing a stronger directional penalty on environment-associated features. This feature-reweighting mechanism reduces reliance on spurious correlations without erasing information, offering a label-free, continuous relaxation of last-layer reweighting methods. Experiments on a controlled synthetic datasets and the DrugOOD datasets show that SCARCE achieves state-of-the-art OOD generalization among label-free methods, using no class labels during representation learning.

cs.LG↗
arXiv2026

Regime-Aware Interpretable Ranking for Sector Rotation

Sector rotation is a common decision in benchmark-relative portfolio management, where portfolio managers rank sectors for overweight and underweight positions. We formulate this task as ranking sector actions by expected future risk-adjusted active performance, measured with forward sector-benchmark Sharpe ratios. Such rankings should preserve temporal information, express model outputs through economically familiar signal characteristics, account for changes in signal relevance across macroeconomic environments, and capture dependencies among sectors. We propose RegimeRank, an interpretable sector-ranking model that represents fundamental and technical signals through expert-defined temporal characteristics, conditions them on learned macro factors via cross-attention, and models cross-sector dependencies through self-attention. Expected-gradient attributions identify the contributions of individual signal-transformation pairs, while macroeconomic and cross-sector attention maps provide diagnostics of the learned conditioning and relational structure. We evaluate RegimeRank on weekly S&P 500 data from 2002 to 2024 under combinatorial purged cross-validation, where it outperforms representative ranking and machine-learning baselines across multiple ranking metrics. Ablation studies support the contributions of latent macro states, regime conditioning, and cross-sector modelling, while attribution and attention analyses illustrate the temporal characteristics, macro states, and sector relationships underlying its rankings.

q-fin.PM↗
arXiv2026

Retrieval-Augmented Diffusion Modeling for Stochastic Discount Factor Portfolios

In this work, we study portfolio optimization under the stochastic discount factor (SDF) framework by learning market state representations that capture the underlying risk structures of financial data. This is challenging due to several factors: financial markets exhibit non-stationary dynamics with shifting regimes, multimodal inputs such as price and news data often contain stochastic noise, and existing diffusion-based approaches, while effective for modeling stochastic dynamics, rely on assumptions such as isotropic Gaussian noise that fail to capture the state-dependent nature of financial uncertainty. To address these challenges, we introduce RADAR, a retrieval-augmented diffusion framework that learns market representations by conditioning on similar historical regimes. RADAR leverages retrieval to construct context-dependent noise distributions, applies conditional diffusion to denoise multimodal representations, and initializes the diffusion process using empirical statistics to reflect state-dependent uncertainty. Experiments show that RADAR achieves state-of-the-art performance on key risk-adjusted metrics while producing economically meaningful signals on asset returns and correlations.

cs.LG↗
arXiv2026

JAREX: An Acquisition Function for Multi-Objective Algorithmic Process Characterization

Pharmaceutical process characterization is central to Quality by Design because it defines how variations in process parameters affect the ability to meet product quality specifications, thereby supporting proven acceptable ranges and robust manufacturing. In practice, however, characterization still relies largely on factorial design of experiments (DOE) approaches, which are inefficient for resolving multivariate pass/fail boundaries in higher-dimensional spaces. While Bayesian optimization has transformed process optimization, adaptive methods for multi-objective process characterization remain lacking. Here, we introduce JAREX (Joint Acceptable Region EXploration), a Bayesian active-learning acquisition function for multi-objective process characterization. JAREX formulates characterization as a joint boundary-learning problem and adaptively selects experiments to recover the joint pass region defined by simultaneous satisfaction of threshold criteria across multiple objectives. JAREX combines an optimistic joint-feasibility mask with a multi-objective extension of randomized straddle, focusing sampling on the joint edge of failure. Our benchmark study suggests that JAREX provides more accurate and sample-efficient recovery of the joint pass region than factorial DOE, space-filling designs, and greedy objective-wise strategies over the full experimental budget range. For batched experimentation, it reduces the number of iterative process characterization experiments by more than half while preserving high accuracy for the boundary-identification task. Implemented in the open-source obsidian package, JAREX provides a modular framework for adaptive, data-efficient multi-objective algorithmic process characterization, supporting sample-efficient range finding in high-dimensional spaces.

stat.ML↗
arXiv2026

Physical Signatures of Supercritical Fluid Boundaries

In the supercritical fluid (SCF) region, at temperatures and pressures above the critical point, the thermodynamic singularity separating liquids and gases no longer exists. Recent arguments based on thermodynamics and critical scalings have revived the proposal that the SCF constitutes an intermediate state of matter, separated from the liquid and gas by two supercritical boundaries, the $L^\pm$ lines. However, until now, the nature of the supercritical state and the physical signatures of these boundaries have remained elusive. Here, we demonstrate that the SCF is characterized by distinct structural, transport, and dynamical behavior. Specifically, the spatial arrangement of particles-captured by the radial distribution function-as well as the diffusion coefficient, shear viscosity, and velocity autocorrelation function in the SCF regime are qualitatively different from those in both the liquid and gas states and exhibit clear physical signatures upon crossing the $L^\pm$ lines. Our theoretical predictions are validated by molecular dynamics simulations of argon and are further supported by existing experimental evidence. These results provide a clear physical foundation for a refined phase diagram of matter in the supercritical region, comprising three distinct states-gas, supercritical fluid, and liquid-separated by two crossover boundaries obeying universal scaling laws.

cond-mat.stat-mech↗
arXiv2026

Toward Robust Single-Photon Perception for Robots: A Condition-Aware Active Learning Approach

LiDAR-based perception plays a fundamental role in modern robotic systems for environment understanding and navigation. Single-photon LiDAR (SPL) extends conventional LiDAR by enabling photon-efficient 3D sensing under challenging conditions such as long-range operation, low-albedo targets, and limited signal returns. However, developing SPL perception models for real-world robotic applications remains difficult because annotated SPL data are costly to obtain and model performance can vary substantially across imaging conditions. In this paper, we present the first active learning framework tailored to the SPL sensing modality rather than a specific downstream task. Our method introduces a physics-grounded, imaging condition-aware sampling strategy that uses synthetic SPL variants to characterize how candidate samples respond to changes in sensing conditions. By jointly modeling prediction uncertainty, sample diversity, and sensitivity to photon-level imaging variations, the proposed approach prioritizes samples that are informative for improving labeling efficiency and robustness. Extensive experiments on synthetic and real-world SPL datasets demonstrate that our method substantially reduces annotation requirements while maintaining strong performance across image-level and dense prediction settings. On synthetic data, our approach achieves 97% classification accuracy using only 1.5% labeled samples. On real-world data, it maintains 90% accuracy with 8.2% labeled samples, outperforming the strongest baseline by 6%. Segmentation results further show that the same condition-aware acquisition principle improves annotation efficiency and robustness across imaging conditions. These results establish a modality-aware active learning strategy for data-efficient SPL perception, with the potential to extend to a broader range of downstream tasks.

eess.IV↗
arXiv2026

TBDub: Production-Oriented Visual Dubbing

Visual dubbing must synchronize mouth motion with replacement speech while preserving identity, appearance, and temporal consistency. Although X-Dub provides a strong mask-free video-editing baseline, its application to livestream and generated-video content reveals limitations in production-domain robustness, temporal and motion stability, identity and oral-detail preservation, and inference efficiency. We present \textbf{TBDub}, a production-oriented extension of X-Dub that combines task-adaptive post-training with task-aware few-step distillation. Post-training adapts the video DiT using production-domain data, production-specific conditioning and filtering, and enhanced audio features to obtain a 30-step Teacher. Distillation adapts DMD/DMD2 to conditional video editing and compresses the Teacher into a two-step Student. On 38 TalkVid clips, the Teacher improves all eight reported reconstruction, perceptual, identity, and synchronization metrics over X-Dub. In the MOS evaluation, it improves lip-sync consistency, identity consistency, and visual quality over X-Dub by 0.14, 0.95, and 0.90 points, while the Student achieves the highest lip-sync and visual-quality scores and remains close to the Teacher in identity consistency. In paired end-to-end generation timing from the first VAE encode through the final VAE decode on a single NVIDIA H20 GPU at $512\times512$, the Student reaches 7.13 effective FPS and reduces total latency by $13.93\times$; the DiT stage alone is accelerated by $42.49\times$. The Student largely retains the Teacher's generation quality and audiovisual synchronization. The code is available on GitHub at \https://github.com/TaoLiveAIGC/TBDub, and the 30-step Teacher and two-step Student weights are available on Hugging Face at https://huggingface.co/TaoLiveAIGC/TBDub.

cs.CV↗
arXiv2026

Variational Probabilistic Quantization for Secret Key Generation

Secret key generation from correlated observations at Alice and Bob, in the presence of an eavesdropper Eve, underpins physical-layer security. Classical pipelines quantize by hand, amplify privacy afterwards, and optimize no objective tied to a key rate. We propose Variational Probabilistic Quantization (VPQ): neural encoders that map the correlated sources directly into a discrete key alphabet, trained by a variational adversarial objective whose entropy, mismatch, and leakage terms match the three terms of the one-way secret key rate. A linear code-offset secure sketch then reconciles the encoder outputs into an identical key without a separate privacy amplification step. We prove that the VPQ losses lower-bound the one-way secret key capacity of the induced source, and derive in closed form the optimal worst-case key rate over the source class of a given alphabet size and mismatch probability, attained by finite-field linear sketches. On Gaussian fading channels, VPQ leaks less to a correlated eavesdropper than one classical and two recent learning-based baselines, and Reed--Solomon reconciliation operates within the predicted finite-blocklength gap.

cs.IT↗
arXiv2026

Universal supercritical thermodynamics for black holes

We investigate thermodynamic crossovers for black holes in the supercritical regime beyond the critical point, where small and large black holes become indistinguishable from the conventional viewpoint. We establish a refined supercritical phase diagram that comprehensively characterizes the phases of small, large, and indistinguishable black holes, delineated by two supercritical crossover lines. The universal scaling laws of these crossover lines are fully verified using the thermodynamics of RN-AdS black holes in both the standard framework and the extended thermodynamic phase space, where the cosmological constant is treated as pressure, as well as in four other black hole systems. Analogies with supercritical crossovers observed in liquid-gas and liquid-liquid phase transitions are discussed. This work can be extended to more complex black hole backgrounds and offers valuable insights into the fundamental nature of black hole thermodynamics.

gr-qc↗
arXiv2026

Supercritical-subcritical correspondence, asymmetric effects and antisymmetric corrections near a critical point

The second-order phase transitions in the Ising model and liquid-gas systems share a universality class and critical exponents, despite the absence of $Z_2$ symmetry in the liquid-gas Hamiltonian. This discrepancy highlights a central puzzle in critical phenomena: what is the influence of asymmetry on scaling laws? For over a century, this question has been explored through examining violations of the empirical ``rectilinear diameter law'' for the subcritical coexistence curve, where asymmetry could generate singular corrections. Here, we extend this investigation to the supercritical regime. We propose a supercritical-subcritical correspondence, drawing a formal analogy between the subcritical coexistence curve and recently defined supercritical boundary lines ($L^\pm$ lines). Our theory predicts that the linear mixing of physical fields - a hallmark of asymmetric systems - produces universal scaling corrections, with antisymmetric coefficients, in these supercritical loci. We verify these predictions using liquid-gas data from the NIST database and a model liquid-liquid transition. Furthermore, we demonstrate that the same asymmetric scaling framework governs the behavior of higher-order cumulants in the order parameter distribution.

cond-mat.stat-mech↗
arXiv2026

Poisson2Gaussian: Noise Gaussianization to Enhance Image Denoising

The quantum nature of light determines the inherent Poisson stochasticity of photon detection, which is ubiquitous in photography, microscopy, and astronomy. However, our controlled numerical studies reveal that the signal-dependency, heteroscedasticity, and statistical asymmetry of Poisson-mixed noise make it challenging for existing denoisers to learn. In contrast, i.i.d. Gaussian noise, with its statistical independence and symmetric distribution, is easier to model for networks. To address this gap, we propose Poisson2Gaussian (P2G), a noise Gaussianization method that explicitly converts complex real-world noise to i.i.d. Gaussian noise via probability density matching beyond low-order moments. We also design an unbiased denoising framework that synergizes P2G with downstream denoisers, ensuring convergence to the underlying signal without requiring paired clean data or explicit noise parameters. Extensive experiments demonstrate that P2G consistently achieves state-of-the-art performance across diverse datasets. In challenging scenarios where noise strongly deviates from Gaussian statistics, our method improves the PSNR by up to 0.75 dB. Notably, P2G is architecture-agnostic and can provide universal improvements for various denoisers. The source code will be publicly available.

cs.CV↗
arXiv2026

SP-TransientBench: A Real-Captured Single Photon Perception Benchmark

Single-photon LiDAR (SPL) based on single-photon avalanche diode (SPAD) sensing enables time-resolved photon measurements with extreme sensitivity, offering unique potential for active 3D perception in photon-starved scenarios.However, real-world single photon perception remains fundamentally challenging due to unique measurement noise and complex multi-return transient phenomena, which jointly complicate geometric reconstruction and semantic scene understanding. Despite growing interest in SPAD-based sensing, existing studies are largely limited to simulated data or small-scale controlled captures. As a result, systematic evaluation of real-world single photon perception across depth estimation, multi-view reconstruction, and 3D semantic understanding remains underexplored. To bridge this gap, we introduce SP-TransientBench (STB), a real-captured multi-task benchmark for single photon perception. SP-TransientBenc comprises 10 diverse scenes and 10,297 views captured using a solid-state single-photon LiDAR at $256\times192$ resolution. Each view provides full time-of-flight histograms with multi-return behavior,standardized metadata, and calibrated camera poses for multi-view evaluation. We further provide 13-class 3D semantic annotations for selected scenes. By providing dedicated data splits and evaluation protocols for each task, STB enables consistent and reproducible benchmarking of real-world single photon perception across multiple 3D vision problems. The dataset and code will be released upon acceptance.

cs.CV↗
arXiv2026

Deeper detection limits in astronomical imaging using self-supervised spatiotemporal denoising

The detection limit of astronomical imaging observations is limited by several noise sources. Some of that noise is correlated between neighbouring image pixels and exposures, so in principle could be learned and corrected. We present an astronomical self-supervised transformer-based denoising algorithm (ASTERIS), that integrates spatiotemporal information across multiple exposures. Benchmarking on mock data indicates that ASTERIS improves detection limits by 1.0 magnitude at 90% completeness and purity, while preserving the point spread function and photometric accuracy. Observational validation using data from the James Webb Space Telescope (JWST) and Subaru telescope identifies previously undetectable features, including low-surface-brightness galaxy structures and gravitationally-lensed arcs. Applied to deep JWST images, ASTERIS identifies three times more redshift > 9 galaxy candidates, with rest-frame ultraviolet luminosity 1.0 magnitude fainter, than previous methods.

astro-ph.IM↗
arXiv2026

Robust Flat Magnetoresistivity in D0$_3$-Fe$_3$Ga Driven by Chiral Anomaly

Topologically non-trivial nodes emerging from flat-band crossings not only enhance unconventional topological responses but also play a fundamental role in exploring correlation-driven topological physics. Here, we report the exceptionally robust chiral-anomaly-dominated transport in D0_3-Fe_3Ga. First, we observe a combination of positive and negative magnetoresistance, ideal planar longitudinal magnetoresistance (PLMR), and the planar Hall effect (PHE). Second, ultra-low-temperature resistivity exhibits pronounced non-Fermi-liquid (NFL) behavior, accompanied by the emergence of giant intrinsic anomalous Hall conductivity (AHC), in excellent agreement with our DFT calculations, which confirm the existence of tilted Weyl points arising from crossings of nearly three-dimensional (3D) flat bands. Most remarkably, we detect an exceptionally robust flat magnetoresistance (flat-MR) that persists without decay up to 33 T. This set of phenomena provides strong evidence that the Fermi level intersects the flattened Weyl crossings, offering confirmation of a topological flat-band semimetal. D0_3-Fe_3Ga presents a promising magnetic platform for quantum device innovations.

cond-mat.mtrl-sci↗