arXiv subjects

Yu Huang

Publications and source records attributed to Yu Huang.

At least 91 records · Page 5Linked to original sources

arXiv2026

Conditional diffusion models for downscaling and bias correction of Earth system model precipitation

Climate change exacerbates extreme weather events like heavy rainfall and flooding. As these events cause severe socioeconomic damage, accurate high-resolution simulation of precipitation is imperative. However, existing Earth System Models (ESMs) struggle to resolve small-scale dynamics and suffer from biases. Traditional statistical bias correction and downscaling methods fall short in improving spatial structure, while recent deep learning methods lack controllability and suffer from unstable training. Here, we propose a machine learning framework for simultaneous bias correction and downscaling. We first map observational and ESM data to a shared embedding space, where both are unbiased towards each other, and then train a conditional diffusion model to reverse the mapping. Only observational data is used for the training, so that the diffusion model can be employed to correct and downscale any ESM field without need for retraining. Our approach ensures statistical fidelity and preserves spatial patterns larger than a chosen spatial correction scale. We demonstrate that our approach outperforms existing statistical and deep learning methods especially regarding extreme events.

physics.geo-ph↗

arXiv2026

Direct Observation of Unidirectional Density Wave and Band splitting in a Single-Domain Trilayer Nickelate Pr$_4$Ni$_3$O$_{10}$

Unraveling the interplay between density-wave (DW) instabilities and multi-orbital physics is critical for understanding superconductivity in Ruddlesden-Popper nickelates, yet intrinsic electronic features have been persistently obscured by material inhomogeneity and thus the multi-domain averaging effect. Here, we employ micro-focused angle-resolved photoemission spectroscopy ($μ$-ARPES) on single-domain Pr$_4$Ni$_3$O$_{10}$ to disentangle the complex hierarchy of intrinsic and back-folded bands, explicitly identifying the electronic states driving the DW phase transition. We provide decisive spectroscopic evidence that the low-energy reconstruction is governed by inter-orbital nesting between the $α$ and $β$ bands. Specifically, we resolve a orbital-dependent gap of $\sim44$ meV on the $α$ pocket, a value quantitatively consistent with prior measurements, unifying previously conflicting experimental reports regarding the locus and magnitude of the DW gap. Furthermore, we reveal strong orbital-selective mass renormalization in the $d_{z^2}$ states and successfully resolve the long-sought intrinsic trilayer $β$-band splitting, establishing a critical lower bound for the outer-layer hopping. These results define a coherent microscopic fingerprint for the trilayer nickelates, identifying the specific nesting channels and correlation effects that underpin the phase diagram.

cond-mat.supr-con↗

arXiv2026

SysMoBench: Evaluating AI on Formally Modeling Complex Real-World Systems

Formal models are essential to specifying large, complex computer systems and verifying their correctness, but are notoriously expensive to write and maintain. Recent advances in generative AI show promise in generating certain forms of specifications. However, existing work mostly targets small code, not complete systems. It is unclear whether AI can deal with realistic system artifacts, as this requires abstracting their complex behavioral properties into formal models. We present SysMoBench, a benchmark that evaluates AI's ability to formally model large, complex systems. We focus on concurrent and distributed systems, which are keystones of today's critical computing infrastructures, encompassing operating systems and cloud infrastructure. We use TLA+, the de facto specification language for concurrent and distributed systems, though the benchmark can be extended to other specification languages. We address the primary challenge of evaluating AI-generated models by automating metrics like syntactic and runtime correctness, conformance to system code, and invariant correctness. SysMoBench currently includes eleven diverse system artifacts: the Raft implementation of Etcd and Redis, the leader election of ZooKeeper, the Spinlock, Mutex, and Ringbuffer in Asterinas OS, etc., with more being added. SysMoBench enables us to understand the capabilities and limitations of today's LLMs and agents, putting tools in this area on a firm footing and opening up promising new research directions.

cs.AI↗

arXiv2026

Preconditioning Benefits of Spectral Orthogonalization in Muon

The Muon optimizer, a matrix-structured algorithm that leverages spectral orthogonalization of gradients, is a milestone in the pretraining of large language models. However, the underlying mechanisms of Muon -- particularly the role of gradient orthogonalization -- remain poorly understood, with very few works providing end-to-end analyses that rigorously explain its advantages in concrete applications. We take a step by studying the effectiveness of a simplified variant of Muon through two case studies: matrix factorization, and in-context learning of linear transformers. For both problems, we prove that simplified Muon converges linearly with iteration complexities independent of the relevant condition number, provably outperforming gradient descent and Adam. Our analysis reveals that the Muon dynamics decouple into a collection of independent scalar sequences in the spectral domain, each exhibiting similar convergence behavior. Our theory formalizes the preconditioning effect induced by spectral orthogonalization, offering insight into Muon's effectiveness in these matrix optimization problems and potentially beyond.

cs.LG↗

arXiv2026

Interaction-induced chiral-transport inversion

We investigate the chiral dynamics of locally interacting bosons in a two-leg flux ladder, where on-site interactions, despite being fully isotropic, counterintuitively reverse the flux-induced chiral transport of density distribution. For a Bose-Einstein condensate (in the mean-field regime), this reversal arises from an interactiondriven dynamical band-occupation inversion, which selectively populates single-particle states of the opposing chirality. Strikingly, the chiral-transport inversion has a few-body, hence beyond-mean-field, origin, as the formation of two-body bound states with reversed chirality dominates the few-body dynamics. This dual pathway, that is, occupation inversion and bound-state formation, underlies the chiral-transport inversion, which challenges the conventional wisdom that isotropic interactions cannot bias density transport. Our work reveals the interplay between interactions and chirality and highlights how correlations engineer exotic quantum transport.

cond-mat.quant-gas↗

arXiv2026

MedReflect: Teaching Medical LLMs to Self-Improve via Reflective Correction

Medical problem-solving demands expert knowledge and intricate reasoning. Recent studies of large language models (LLMs) attempt to ease this complexity by introducing external knowledge verification through retrieval-augmented generation or by training on reasoning datasets. However, these approaches suffer from drawbacks such as retrieval overhead and high annotation costs, and they heavily rely on substituted external assistants to reach limited performance in medical field. In this paper, we introduce MedReflect, a generalizable framework designed to inspire LLMs with a physician-like reflective thinking mode. MedReflect generates a single-pass reflection chain that includes initial hypothesis generation, self-questioning, self-answering and decision refinement. This self-verified and self-reflective nature releases large language model's latent capability in medical problem-solving without external retrieval or heavy annotation. We demonstrate that MedReflect enables cost-efficient medical dataset construction. With only a minimal subset of randomly sampled training examples and lightweight fine-tuning, this approach achieves notable absolute accuracy improvements across a series of medical benchmarks while significantly cutting annotation requirements. Our results provide evidence that LLMs can learn to solve specialized medical problems via self-reflection and self-improvement, reducing reliance on external supervision and extensive task-specific fine-tuning data.

cs.CL↗

arXiv2026

XLinear: A Lightweight and Accurate MLP-Based Model for Long-Term Time Series Forecasting with Exogenous Inputs

Despite the prevalent assumption of uniform variable importance in long-term time series forecasting models, real world applications often exhibit asymmetric causal relationships and varying data acquisition costs. Specifically, cost-effective exogenous data (e.g., local weather) can unilaterally influence dynamics of endogenous variables, such as lake surface temperature. Exploiting these links enables more effective forecasts when exogenous inputs are readily available. Transformer-based models capture long-range dependencies but incur high computation and suffer from permutation invariance. Patch-based variants improve efficiency yet can miss local temporal patterns. To efficiently exploit informative signals across both the temporal dimension and relevant exogenous variables, this study proposes XLinear, a lightweight time series forecasting model built upon MultiLayer Perceptrons (MLPs). XLinear uses a global token derived from an endogenous variable as a pivotal hub for interacting with exogenous variables, and employs MLPs with sigmoid activation to extract both temporal patterns and variate-wise dependencies. Its prediction head then integrates these signals to forecast the endogenous series. We evaluate XLinear on seven standard benchmarks and five real-world datasets with exogenous inputs. Compared with state-of-the-art models, XLinear delivers superior accuracy and efficiency for both multivariate forecasts and univariate forecasts influenced by exogenous inputs.

cs.LG↗

arXiv2026

A Visual Semantic Adaptive Watermark grounded by Prefix-Tuning for Large Vision-Language Model

Watermarking has emerged as a pivotal solution for content traceability and intellectual property protection in Large Vision-Language Models (LVLMs). However, vision-agnostic watermarks introduce visually irrelevant tokens and disrupt visual grounding by enforcing indiscriminate pseudo-random biases, while some semantic-aware methods incur prohibitive inference latency due to rejection sampling. In this paper, we propose the VIsual Semantic Adaptive Watermark (VISA-Mark), a novel framework that embeds detectable signals while strictly preserving visual fidelity. Our approach employs a lightweight, efficiently trained prefix-tuner to extract dynamic Visual-Evidence Weights, which quantify the evidentiary support for candidate tokens based on the visual input. These weights guide an adaptive vocabulary partitioning and logits perturbation mechanism, concentrating watermark strength specifically on visually-supported tokens. By actively aligning the watermark with visual evidence, VISA-Mark effectively maintains visual fidelity. Empirical results confirm that VISA-Mark outperforms conventional methods with a 7.8% improvement in visual consistency (Chair-I) and superior semantic fidelity. The framework maintains highly competitive detection accuracy (96.88% AUC) and robust attack resilience (99.3%) without sacrificing inference efficiency, effectively establishing a new standard for reliability-preserving multimodal watermarking.

cs.CV↗

arXiv2026

VisionReward: Fine-Grained Multi-Dimensional Human Preference Learning for Image and Video Generation

Visual generative models have achieved remarkable progress in synthesizing photorealistic images and videos, yet aligning their outputs with human preferences across critical dimensions remains a persistent challenge. Though reinforcement learning from human feedback offers promise for preference alignment, existing reward models for visual generation face limitations, including black-box scoring without interpretability and potentially resultant unexpected biases. We present VisionReward, a general framework for learning human visual preferences in both image and video generation. Specifically, we employ a hierarchical visual assessment framework to capture fine-grained human preferences, and leverages linear weighting to enable interpretable preference learning. Furthermore, we propose a multi-dimensional consistent strategy when using VisionReward as a reward model during preference optimization for visual generation. Experiments show that VisionReward can significantly outperform existing image and video reward models on both machine metrics and human evaluation. Notably, VisionReward surpasses VideoScore by 17.2% in preference prediction accuracy, and text-to-video models with VisionReward achieve a 31.6% higher pairwise win rate compared to the same models using VideoScore. All code and datasets are provided at https://github.com/THUDM/VisionReward.

cs.CV↗

arXiv2025

Capillary Filling Dynamics in Polygonal Tubes

We study the dynamics of capillary filling in tubes of regular polygon cross-section. Using Onsager variational principle, we derive a coupled ordinary differential equation and partial differential equation, which respectively describe time evolution of the bulk flow and the saturation profile of the finger flow. We obtain both numerical solution and self-similar solution to the coupled equations, and the results indicate that the bulk flow and the finger flow both follow the $t^{1/2}$ time-scaling. We show that due to the coupling effect of the finger flow, the prefactor for the bulk flow is smaller than that of the Lucas-Washburn prediction. The reduction effect is more pronounced when the side number $n$ of the regular-polygon is small, while as $n$ increases, the prefactor approaches Lucas-Washburn prediction.

physics.flu-dyn↗

arXiv2025

$^{12}$CO $J$=3--2 Observations of Tycho's supernova remnant: constraints on the environmental gas properties

Recent observations suggest that Tycho's supernova remnant (SNR; SN 1572) is expanding into a cavity wall of molecular clouds (MCs), which decelerate the SNR and influence its multi-wavelength morphology. To constrain the physical properties of environmental MCs and search for heated gas, we perform a JCMT $^{12}$CO $J$=3--2 observation and compare with previous $^{12}$CO $J$=2--1, $^{12}$CO $J$=1--0 and $^{13}$CO $J$=1--0 data. We present the $^{12}$CO $J$=3--2 map toward Tycho and show that the $^{12}$CO $J$=3--2 spatial distribution and line profiles are similar to those of the lower-$J$ CO lines. By comparing the multiple transitions of CO and the RADEX (Radiative transfer code in non-Local Thermodynamic Equilibrium) models, we constrain the physical properties of molecular gas surrounding Tycho: the northern cloud has a molecular column density of $N({\rm H}_{2})=0.5$ -- $4.5\times 10^{22}$ cm$^{-2}$, while other regions have $N({\rm H}_{2})=0.2$ -- $3.9\times10^{21}$ cm$^{-2}$; the kinetic temperatures $T_{\rm k}$ of these clouds are in the range of 9 -- 22 K and the volume densities $n({\rm H}_{2})$ are 20 -- $700$ cm$^{-3}$. We also discuss the difficulty in finding hot molecular gas shocked by such a young SNR. We estimate that the shocked molecular layer can be as thin as 0.003 pc, corresponding to $0.2''$ at the distance of 2.5 kpc, which is 2 orders of magnitude smaller than the angular resolution of current CO observations. Therefore, our molecular observations are largely insensitive to the thin shocked gas layer; instead, they detect the environmental gas.

astro-ph.GA↗

arXiv2025

Multi-head Transformers Provably Learn Symbolic Multi-step Reasoning via Gradient Descent

Transformers have demonstrated remarkable capabilities in multi-step reasoning tasks. However, understandings of the underlying mechanisms by which they acquire these abilities through training remain limited, particularly from a theoretical standpoint. This work investigates how transformers learn to solve symbolic multi-step reasoning problems through chain-of-thought processes, focusing on path-finding in trees. We analyze two intertwined tasks: a backward reasoning task, where the model outputs a path from a goal node to the root, and a more complex forward reasoning task, where the model implements two-stage reasoning by first identifying the goal-to-root path and then reversing it to produce the root-to-goal path. Our theoretical analysis, grounded in the dynamics of gradient descent, shows that trained one-layer transformers can provably solve both tasks with generalization guarantees to unseen trees. In particular, our multi-phase training dynamics for forward reasoning elucidate how different attention heads learn to specialize and coordinate autonomously to solve the two subtasks in a single autoregressive path. These results provide a mechanistic explanation of how trained transformers can implement sequential algorithmic procedures. Moreover, they offer insights into the emergence of reasoning abilities, suggesting that when tasks are structured to take intermediate chain-of-thought steps, even shallow multi-head transformers can effectively solve problems that would otherwise require deeper architectures.

cs.LG↗

arXiv2025

Learning to Equalize: Data-Driven Frequency-Domain Signal Recovery in Molecular Communications

In molecular communications (MC), inter-symbol interference (ISI) and noise are key factors that degrade communication reliability. Although time-domain equalization can effectively mitigate these effects, it often entails high computational complexity concerning the channel memory. In contrast, frequency-domain equalization (FDE) offers greater computational efficiency but typically requires prior knowledge of the channel model. To address this limitation, this letter proposes FDE techniques based on long short-term memory (LSTM) neural networks, enabling temporal correlation modeling in MC channels to improve ISI and noise suppression. To eliminate the reliance on prior channel information in conventional FDE methods, a supervised training strategy is employed for channel-adaptive equalization. Simulation results demonstrate that the proposed LSTM-FDE significantly reduces the bit error rate compared to traditional FDE and feedforward neural network-based equalizers. This performance gain is attributed to the LSTM's temporal modeling capabilities, which enhance noise suppression and accelerate model convergence, while maintaining comparable computational efficiency.

q-bio.SC↗

arXiv2025

VideoMark: A Distortion-Free Robust Watermarking Framework for Video Diffusion Models

This work introduces \textbf{VideoMark}, a distortion-free robust watermarking framework for video diffusion models. As diffusion models excel in generating realistic videos, reliable content attribution is increasingly critical. However, existing video watermarking methods often introduce distortion by altering the initial distribution of diffusion variables and are vulnerable to temporal attacks, such as frame deletion, due to variable video lengths. VideoMark addresses these challenges by employing a \textbf{pure pseudorandom initialization} to embed watermarks, avoiding distortion while ensuring uniform noise distribution in the latent space to preserve generation quality. To enhance robustness, we adopt a frame-wise watermarking strategy with pseudorandom error correction (PRC) codes, using a fixed watermark sequence with randomly selected starting indices for each video. For watermark extraction, we propose a Temporal Matching Module (TMM) that leverages edit distance to align decoded messages with the original watermark sequence, ensuring resilience against temporal attacks. Experimental results show that VideoMark achieves higher decoding accuracy than existing methods while maintaining video quality comparable to watermark-free generation. The watermark remains imperceptible to attackers without the secret key, offering superior invisibility compared to other frameworks. VideoMark provides a practical, training-free solution for content attribution in diffusion-based video generation. Our code and data are available at \href{https://github.com/KYRIE-LI11/VideoMark}{VideoMark}.

cs.CR↗

arXiv2025

Video Signature: Implicit Watermarking for Video Diffusion Models

The rapid development of Artificial Intelligence Generated Content (AIGC) has led to significant progress in video generation, but also raises serious concerns about intellectual property protection and reliable content tracing. Watermarking is a widely adopted solution to this issue, yet existing methods for video generation mainly follow a post-generation paradigm, which often fails to effectively balance the trade-off between video quality and watermark extraction. Meanwhile, current in-generation methods that embed the watermark into the initial Gaussian noise usually incur substantial additional computation. To address these issues, we propose \textbf{Video Signature} (\textsc{VidSig}), an implicit watermarking method for video diffusion models that enables imperceptible and adaptive watermark integration during video generation with almost no extra latency. Specifically, we partially fine-tune the latent decoder, where \textbf{Perturbation-Aware Suppression} (PAS) pre-identifies and freezes perceptually sensitive layers to preserve visual quality. Beyond spatial fidelity, we further enhance temporal consistency by introducing a lightweight \textbf{Temporal Alignment} module that guides the decoder to generate coherent frame sequences during fine-tuning. Experimental results show that \textsc{VidSig} achieves the best trade-off among watermark extraction accuracy, video quality, and watermark latency. It also demonstrates strong robustness against both spatial and temporal tamper, and remains stable across different video lengths and resolutions, highlighting its practicality in real-world scenarios.

cs.CV↗

arXiv2025

AnomalyAID: Reliable Interpretation for Semi-supervised Network Anomaly Detection

Semi-supervised Learning plays a crucial role in network anomaly detection applications, however, learning anomaly patterns with limited labeled samples is not easy. Additionally, the lack of interpretability creates key barriers to the adoption of semi-supervised frameworks in practice. Most existing interpretation methods are developed for supervised/unsupervised frameworks or non-security domains and fail to provide reliable interpretations. In this paper, we propose AnomalyAID, a general framework aiming to (1) make the anomaly detection process interpretable and improve the reliability of interpretation results, and (2) assign high-confidence pseudo labels to unlabeled samples for improving the performance of anomaly detection systems with limited supervised data. For (1), we propose a novel interpretation approach that leverages global and local interpreters to provide reliable explanations, while for (2), we design a new two-stage semi-supervised learning framework for network anomaly detection by aligning both stages' model predictions with special constraints. We apply AnomalyAID over two representative network anomaly detection tasks and extensively evaluate AnomalyAID with representative prior works. Experimental results demonstrate that AnomalyAID can provide accurate detection results with reliable interpretations for semi-supervised network anomaly detection systems.

cs.LG↗

arXiv2025

Transformers Provably Learn Chain-of-Thought Reasoning with Length Generalization

The ability to reason lies at the core of artificial intelligence (AI), and challenging problems usually call for deeper and longer reasoning to tackle. A crucial question about AI reasoning is whether models can extrapolate learned reasoning patterns to solve harder tasks with longer chain-of-thought (CoT). In this work, we present a theoretical analysis of transformers learning on synthetic state-tracking tasks with gradient descent. We mathematically prove how the algebraic structure of state-tracking problems governs the degree of extrapolation of the learned CoT. Specifically, our theory characterizes the length generalization of transformers through the mechanism of attention concentration, linking the retrieval robustness of the attention layer to the state-tracking task structure of long-context reasoning. Moreover, for transformers with limited reasoning length, we prove that a recursive self-training scheme can progressively extend the range of solvable problem lengths. To our knowledge, we provide the first optimization guarantee that constant-depth transformers provably learn $\mathsf{NC}^1$-complete problems with CoT, significantly going beyond prior art confined in $\mathsf{TC}^0$, unless the widely held conjecture $\mathsf{TC}^0 \neq \mathsf{NC}^1$ fails. Finally, we present a broad set of experiments supporting our theoretical results, confirming the length generalization behaviors and the mechanism of attention concentration.

cs.LG↗

arXiv2025

Transformers Provably Learn Directed Acyclic Graphs via Kernel-Guided Mutual Information

Uncovering hidden graph structures underlying real-world data is a critical challenge with broad applications across scientific domains. Recently, transformer-based models leveraging the attention mechanism have demonstrated strong empirical success in capturing complex dependencies within graphs. However, the theoretical understanding of their training dynamics has been limited to tree-like graphs, where each node depends on a single parent. Extending provable guarantees to more general directed acyclic graphs (DAGs) -- which involve multiple parents per node -- remains challenging, primarily due to the difficulty in designing training objectives that enable different attention heads to separately learn multiple different parent relationships. In this work, we address this problem by introducing a novel information-theoretic metric: the kernel-guided mutual information (KG-MI), based on the $f$-divergence. Our objective combines KG-MI with a multi-head attention framework, where each head is associated with a distinct marginal transition kernel to model diverse parent-child dependencies effectively. We prove that, given sequences generated by a $K$-parent DAG, training a single-layer, multi-head transformer via gradient ascent converges to the global optimum in polynomial time. Furthermore, we characterize the attention score patterns at convergence. In addition, when particularizing the $f$-divergence to the KL divergence, the learned attention scores accurately reflect the ground-truth adjacency matrix, thereby provably recovering the underlying graph structure. Experimental results validate our theoretical findings.

cs.LG↗