Search arXiv⌕ Search

arXiv subjects

Yibo Wang

Publications and source records attributed to Yibo Wang.

At least 91 records · Page 5Linked to original sources
arXiv2026

QUITE: A Query Rewrite System Beyond Rules with LLM Agents

Query rewrite transforms SQL queries into semantically equivalent forms that run more efficiently. Existing approaches mainly rely on predefined rewrite rules, but they handle a limited subset of queries and can cause performance regressions. This limitation stems from three challenges of rule-based query rewrite: (1) it is hard to discover and verify new rules, (2) fixed rewrite rules do not generalize to new query patterns, and (3) some rewrite techniques cannot be expressed as fixed rules. Motivated by the fact that human experts exhibit significantly better rewrite ability but suffer from scalability, and Large Language Models (LLMs) have demonstrated nearly human-level semantic and reasoning abilities, we propose a new approach of using LLMs to rewrite SQL queries beyond rules. Due to the hallucination problems in LLMs, directly applying LLMs often leads to nonequivalent and suboptimal queries. To address this issue, we propose QUITE (query rewrite), a training-free and feedback-aware system based on LLM agents that rewrites SQL queries into semantically equivalent forms with significantly better performance, covering a broader range of query patterns and rewrite strategies compared to rule-based methods. Firstly, we design a multi-agent framework controlled by a finite state machine (FSM) to equip LLMs with the ability to use external tools and enhance the rewrite process with real-time database feedback. Secondly, we develop a rewrite middleware to enhance the ability of LLMs to generate optimized query equivalents. Finally, we employ a novel hint injection technique to improve execution plans for rewritten queries. Extensive experiments show that QUITE reduces query execution time by up to 35.8% over state-of-the-art approaches and produces 24.1% more rewrites than prior methods, covering query cases that earlier systems did not handle.

cs.DB↗
arXiv2025

Detecting Hallucinations in Graph Retrieval-Augmented Generation via Attention Patterns and Semantic Alignment

Graph-based Retrieval-Augmented Generation (GraphRAG) enhances Large Language Models (LLMs) by incorporating external knowledge from linearized subgraphs retrieved from knowledge graphs. However, LLMs struggle to interpret the relational and topological information in these inputs, resulting in hallucinations that are inconsistent with the retrieved knowledge. To analyze how LLMs attend to and retain structured knowledge during generation, we propose two lightweight interpretability metrics: Path Reliance Degree (PRD), which measures over-reliance on shortest-path triples, and Semantic Alignment Score (SAS), which assesses how well the model's internal representations align with the retrieved knowledge. Through empirical analysis on a knowledge-based QA task, we identify failure patterns associated with over-reliance on salient paths and weak semantic grounding, as indicated by high PRD and low SAS scores. We further develop a lightweight post-hoc hallucination detector, Graph Grounding and Alignment (GGA), which outperforms strong semantic and confidence-based baselines across AUC and F1. By grounding hallucination analysis in mechanistic interpretability, our work offers insights into how structural limitations in LLMs contribute to hallucinations, informing the design of more reliable GraphRAG systems in the future.

cs.CL↗
arXiv2025

Ergodicity and invariant measure approximation of the stochastic Cahn-Hilliard equation via an explicit fully discrete scheme

This paper investigates the stochastic Cahn-Hilliard equation (SCHE) driven by additive space-time white noise. We first refine the analytical ergodic theory by proving that the continuum equation admits a unique invariant measure in the more regular state space H_α, extending the classical result of Da Prato and Debussche (1996) on the negative Sobolev space $\dot{H}^{-1}_α$. To approximate long-time behaviour, we introduce an explicit fully discrete scheme that combines a finite-difference spatial discretization with a strongly tamed exponential Euler method in time. Uniform-in-time moment bounds in the $L^\infty$-norm are established for the numerical solution, and a uniform strong convergence estimate with an explicit rate is derived for the fully discrete approximation. Exploiting a mass-preserving minorization tailored to Neumann boundary conditions, we further show that the numerical scheme is geometrically ergodic and possesses a unique invariant measure, together with polynomial-order error bounds for approximating the exact invariant measure. Strong laws of large numbers are proved for both the continuous and discrete systems, ensuring almost-sure convergence of temporal averages to the corresponding ergodic limits. Numerical experiments corroborate the theoretical findings, including the long-time strong convergence and the accuracy of invariant measure approximation. Overall, the results provide a complete analytical and numerical framework for investigating the long-time statistical behaviour of the SCHE.

math.NA↗
arXiv2025

SPACE: Noise Contrastive Estimation Stabilizes Self-Play Fine-Tuning for Large Language Models

Self-play fine-tuning has demonstrated promising abilities in adapting large language models (LLMs) to downstream tasks with limited real-world data. The basic principle is to iteratively refine the model with real samples and synthetic ones generated from itself. However, the existing methods primarily focus on the relative gaps between the rewards for two types of data, neglecting their absolute values. Through theoretical analysis, we identify that the gap-based methods suffer from unstable evolution, due to the potentially degenerated objectives. To address this limitation, we introduce a novel self-play fine-tuning method, namely Self-PlAy via Noise Contrastive Estimation (SPACE), which leverages noise contrastive estimation to capture the real-world data distribution. Specifically, SPACE treats synthetic samples as auxiliary components, and discriminates them from the real ones in a binary classification manner. As a result, SPACE independently optimizes the absolute reward values for each type of data, ensuring a consistently meaningful objective and thereby avoiding the instability issue. Theoretically, we show that the optimal solution of the objective in SPACE aligns with the underlying distribution of real-world data, and SPACE guarantees a provably stable convergence to the optimal distribution. Empirically, we show that SPACE significantly improves the performance of LLMs over various tasks, and outperforms supervised fine-tuning that employs much more real-world samples. Compared to gap-based self-play fine-tuning methods, SPACE exhibits remarkable superiority and stable evolution.

cs.LG↗
arXiv2025

The Faintest, Extremely Variable X-ray Tidal Disruption Event from a Supermassive Black Hole Binary?

Tidal disruption events (TDEs), which occur when stars enter the tidal radii of supermassive black holes (SMBHs) and are subsequently torn apart by their tidal forces, represent intriguing phenomena that stimulate growing research interest and pose an increasing number of puzzles in the era of time-domain astronomy. Here we report an unusual X-ray transient, XID 935, discovered in the 7 Ms Chandra Deep Field-South, the deepest X-ray survey ever. XID 935 experienced an overall X-ray dimming by a factor of more than 40 between 1999 and 2016. Not monotonically decreasing during this period, its X-ray luminosity increased by a factor $> 27$ within 2 months, from $L_{\rm 0.5-7\ keV}<10^{40.87}$ erg s$^{-1}$ (10 October 2014 -- 4 January 2015) to $L_{\rm 0.5-7\ keV}=10^{42.31\pm 0.20}$ erg s$^{-1}$ (16 March 2015). The X-ray position of XID 935 is located at the center of its host galaxy with a spectroscopic redshift of 0.251, whose optical spectra do not display emission characteristics associated with an active galactic nucleus. The peak 0.5--2.0 keV flux is the faintest among all the X-ray-selected TDE candidates to date. Thanks to a total exposure of $\sim 9.5$ Ms in the X-ray bands, we manage to secure relatively well-sampled, 20-year-long X-ray light curves of this deepest X-ray-selected TDE candidate. We find that a partial TDE model could not explain the main declining trend. An SMBH binary TDE model is in acceptable accordance with the light curves of XID 935; however, it fails to match short-timescale fluctuations exactly. Therefore, the exceptional observational features of XID 935 provide a key benchmark for refining quantitative TDE models and simulations.

astro-ph.HE↗
arXiv2025

Ultraviolet Spectral Evidence for Ansky as a Slowly Evolving Featureless Tidal Disruption Event with Quasiperiodic Eruptions

X-ray quasi-periodic eruptions (QPEs) are rare and enigmatic phenomena that increasingly show a connection to tidal disruption events (TDEs). However, the recently discovered QPEs in ZTF19acnskyy ("Ansky") appear to be linked to an active galactic nucleus (AGN) rather than a TDE, as their slow decay and AGN-like variability differ markedly from that of typical TDEs. This finding may imply broader formation channels for QPEs. To further investigate Ansky's nature, we obtained a timely ultraviolet (UV) spectrum, which reveals a featureless, TDE-like spectrum devoid of broad optical or UV emission lines. Additionally, the steep UV continuum, fitted by a power law with an index of -2.6, aligns more closely with TDEs than with AGNs. Compared to other featureless TDEs, Ansky exhibits a significantly lower blackbody luminosity (10^43 erg s^-1) and much longer rise and decay timescales, suggesting a distinct TDE subclass. An offset TDE involving an intermediate-mass black hole is unlikely, given its position consistent with the galactic center with a 3 sigma upper limit of 54 pc. Instead, we propose that Ansky may result from the tidal disruption of a post-main-sequence star by a typical supermassive black hole. Our findings strengthen the growing evidence for TDE-QPE associations, although other formation channels for QPEs remain plausible and await future observational efforts.

astro-ph.HE↗
arXiv2025

Signatures of magnetism in zigzag graphene nanoribbon embedded in h-BN lattice

Zigzag edges of graphene have long been predicted to exhibit magnetic electronic state near the Fermi level, which can cause spin-related phenomena and offer unique potentials for graphene-based spintronics. However, the magnetic conduction channels along these edges have yet been reported experimentally. Here, we report the observation on signatures of magnetism in zigzag graphene nanoribbons (zGNRs) embedded in hexagonal boron nitride (h-BN). The in-plane bonding with BN can stabilize the edges of zGNRs, and thus enable a direct probing of the intrinsic magnetism. Firstly, the presence of magnetism of a zGNR was confirmed by scanning NV center microscopy. And then, zGNR was fabricated into a transistor with a width of ~9 nm wide and a channel length of sub-50 nm. By performing magneto-transport measurements, Fabry-Pérot interference patterns were observed in the transistor at 4 Kelvin, which indicates a coherent transport through the channel. A large magnetoresistance of ~175 Ω, corresponding to a ratio of ~1.3 %, was observed at the same temperature. More importantly, such magneto-transport signal is highly anisotropic on the magnetic field direction, and its appearance extends well above room temperature. All these evidences corroborate the existence of robust magnetic ordering in the edge state of zGNR. The findings on zGNR embedded in h-BN provide an effective platform for the future exploration of graphene-based spintronic devices.

cond-mat.mes-hall↗
arXiv2025

R1-Compress: Long Chain-of-Thought Compression via Chunk Compression and Search

Chain-of-Thought (CoT) reasoning enhances large language models (LLMs) by enabling step-by-step problem-solving, yet its extension to Long-CoT introduces substantial computational overhead due to increased token length. Existing compression approaches -- instance-level and token-level -- either sacrifice essential local reasoning signals like reflection or yield incoherent outputs. To address these limitations, we propose R1-Compress, a two-stage chunk-level compression framework that preserves both local information and coherence. Our method segments Long-CoT into manageable chunks, applies LLM-driven inner-chunk compression, and employs an inter-chunk search mechanism to select the short and coherent sequence. Experiments on Qwen2.5-Instruct models across MATH500, AIME24, and GPQA-Diamond demonstrate that R1-Compress significantly reduces token usage while maintaining comparable reasoning accuracy. On MATH500, R1-Compress achieves an accuracy of 92.4%, with only a 0.6% drop compared to the Long-CoT baseline, while reducing token usage by about 20%. Source code will be available at https://github.com/w-yibo/R1-Compress

cs.CL↗
arXiv2025

RECODE-H: A Benchmark for Research Code Development with Interactive Human Feedback

Large language models (LLMs) show the promise in supporting scientific research implementation, yet their ability to generate correct and executable code remains limited. Existing works largely adopt one-shot settings, ignoring the iterative and feedback-driven nature of realistic workflows of scientific research development. To address this gap, we present RECODE-H, a benchmark of 102 tasks from research papers and repositories that evaluates LLM agents through multi-turn interactions with LLM-simulated human feedback. It includes structured instructions,unit tests, and a five-level feedback hierarchy to reflect realistic researcher-agent collaboration. We further present ReCodeAgent, a framework that integrates feedback into iterative code generation. Experiments with leading LLMs, including GPT-5, Claude-Sonnet-4, DeepSeek-V3.1, and Gemini 2.5, show substantial performance gains with richer feedback, while also highlighting ongoing challenges in the generation of complex research code. RECODE-H establishes a foundation for developing adaptive, feedback-driven LLM agents in scientific research implementation

cs.CL↗
arXiv2025

A Survey on Agentic Multimodal Large Language Models

With the recent emergence of revolutionary autonomous agentic systems, research community is witnessing a significant shift from traditional static, passive, and domain-specific AI agents toward more dynamic, proactive, and generalizable agentic AI. Motivated by the growing interest in agentic AI and its potential trajectory toward AGI, we present a comprehensive survey on Agentic Multimodal Large Language Models (Agentic MLLMs). In this survey, we explore the emerging paradigm of agentic MLLMs, delineating their conceptual foundations and distinguishing characteristics from conventional MLLM-based agents. We establish a conceptual framework that organizes agentic MLLMs along three fundamental dimensions: (i) Agentic internal intelligence functions as the system's commander, enabling accurate long-horizon planning through reasoning, reflection, and memory; (ii) Agentic external tool invocation, whereby models proactively use various external tools to extend their problem-solving capabilities beyond their intrinsic knowledge; and (iii) Agentic environment interaction further situates models within virtual or physical environments, allowing them to take actions, adapt strategies, and sustain goal-directed behavior in dynamic real-world scenarios. To further accelerate research in this area for the community, we compile open-source training frameworks, training and evaluation datasets for developing agentic MLLMs. Finally, we review the downstream applications of agentic MLLMs and outline future research directions for this rapidly evolving field. To continuously track developments in this rapidly evolving field, we will also actively update a public repository at https://github.com/HJYao00/Awesome-Agentic-MLLMs.

cs.CV↗
arXiv2025

Towards Generalizable Context-aware Anomaly Detection: A Large-scale Benchmark in Cloud Environments

Anomaly detection in cloud environments remains both critical and challenging. Existing context-level benchmarks typically focus on either metrics or logs and often lack reliable annotation, while most detection methods emphasize point anomalies within a single modality, overlooking contextual signals and limiting real-world applicability. Constructing a benchmark for context anomalies that combines metrics and logs is inherently difficult: reproducing anomalous scenarios on real servers is often infeasible or potentially harmful, while generating synthetic data introduces the additional challenge of maintaining cross-modal consistency. We introduce CloudAnoBench, a large-scale benchmark for context anomalies in cloud environments, comprising 28 anomalous scenarios and 16 deceptive normal scenarios, with 1,252 labeled cases and roughly 200,000 log and metric entries. Compared with prior benchmarks, CloudAnoBench exhibits higher ambiguity and greater difficulty, on which both prior machine learning methods and vanilla LLM prompting perform poorly. To demonstrate its utility, we further propose CloudAnoAgent, an LLM-based agent enhanced by symbolic verification that integrates metrics and logs. This agent system achieves substantial improvements in both anomaly detection and scenario identification on CloudAnoBench, and shows strong generalization to existing datasets. Together, CloudAnoBench and CloudAnoAgent lay the groundwork for advancing context-aware anomaly detection in cloud systems. Project Page: https://jayzou3773.github.io/cloudanobench-agent/

cs.AI↗
arXiv2025

Outflow-cloud interaction as the possible origin of the peculiar radio emission in the tidal disruption event AT2018cqh

AT2018cqh is a unique optical tidal disruption event (TDE) discovered in a dwarf galaxy exhibiting delayed X-ray and radio flares. We present the results from high-resolution VLBA and e-MERLIN radio observations of AT2018cqh extending to $δ$t $\sim$ 2250 days post discovery, which reveal a compact radio emission, unresolved at a scale of <~ 0.13 pc at 7.6 GHz, with a high brightness temperature of $T_b$ ~> 4.03 $\times$ 10$^{9}$ K. The radio spectral energy distribution (SED) is found to gradually shift towards a higher peak flux density and frequency over a period of $\sim$1000 days. An equipartition analysis suggests that there is a little change in the radio emitting region over this period, while the electron density increases by a factor of 3. The radio light curve at 0.89 GHz continues to rise, with a bump feature lasting for 240 days. These properties are in contrast to the predictions of standard shockwave model from a diffuse circumnuclear medium, but could be explained if dense clouds exist in the circumnuclear environment. The latter scenario is supported by our hydrodynamic simulations of the interaction of TDE outflow with a cloud, which can reproduce the temporal evolution in the radio SED. This work highlights the importance of the outflow-cloud interaction in explaining the delayed, fast-rising radio emission observed in some TDEs, especially those occurring in galaxies with pre-existing AGN activity.

astro-ph.HE↗
arXiv2025

UltraHorizon: Benchmarking Agent Capabilities in Ultra Long-Horizon Scenarios

Autonomous agents have recently achieved remarkable progress across diverse domains, yet most evaluations focus on short-horizon, fully observable tasks. In contrast, many critical real-world tasks, such as large-scale software development, commercial investment, and scientific discovery, unfold in long-horizon and partially observable scenarios where success hinges on sustained reasoning, planning, memory management, and tool use. Existing benchmarks rarely capture these long-horizon challenges, leaving a gap in systematic evaluation. To bridge this gap, we introduce \textbf{UltraHorizon} a novel benchmark that measures the foundational capabilities essential for complex real-world challenges. We use exploration as a unifying task across three distinct environments to validate these core competencies. Agents are designed in long-horizon discovery tasks where they must iteratively uncover hidden rules through sustained reasoning, planning, memory and tools management, and interaction with environments. Under the heaviest scale setting, trajectories average \textbf{200k+} tokens and \textbf{400+} tool calls, whereas in standard configurations they still exceed \textbf{35k} tokens and involve more than \textbf{60} tool calls on average. Our extensive experiments reveal that LLM-agents consistently underperform in these settings, whereas human participants achieve higher scores, underscoring a persistent gap in agents' long-horizon abilities. We also observe that simple scaling fails in our task. To better illustrate the failure of agents, we conduct an in-depth analysis of collected trajectories. We identify eight types of errors and attribute them to two primary causes: in-context locking and functional fundamental capability gaps. \href{https://github.com/StarDewXXX/UltraHorizon}{Our code will be available here.}

cs.AI↗
arXiv2025

Generalized Correlation Regression for Disentangling Dependence in Clustered Data

Clustered and longitudinal data are pervasive in scientific studies, from prenatal health programs to clinical trials and public health surveillance. Such data often involve non-Gaussian responses--including binary, categorical, and count outcomes--that exhibit complex correlation structures driven by multilevel clustering, covariates, over-dispersion, or zero inflation. Conventional approaches such as mixed-effects models and generalized estimating equations (GEEs) can capture some of these dependencies, but they are often too rigid or impose restrictive assumptions that limit interpretability and predictive performance. We investigate \emph{generalized correlation regression} (GCR), a unified framework that models correlations directly as functions of interpretable covariates while simultaneously estimating marginal means. By applying a generalized $z$-transformation, GCR guarantees valid correlation matrices, accommodates unbalanced cluster sizes, and flexibly incorporates covariates such as time, space, or group membership into the dependence structure. Through applications to modern prenatal care, a longitudinal toenail infection trial, and clustered health count data, we show that GCR not only achieves superior predictive performance over standard methods, but also reveals family-, community-, and individual-level drivers of dependence that are obscured under conventional modeling. These results demonstrate the broad applied value of GCR for analyzing binary, count, and categorical data in clustered and longitudinal settings.

stat.ME↗
arXiv2025

Random attractor and SRB measure for stochastic Hopf bifurcation under discretization

Chaotic phases in stochastic differential equations are characterized by two essential long-time dynamical features: a random attractor capturing asymptotic geometry and a Sinai-Ruelle-Bowen (SRB) measure describing statistical information. This paper investigates whether the stochastic Hopf bifurcation under discretization could inherit both features. We establish that the stochastic Hopf bifurcation under discretization induces a discrete random dynamical system. Further, we prove that this discrete system possesses a random attractor, and then derive the existence of an SRB measure by demonstrating a strictly positive numerical Lyapunov exponent. Numerical experiments visualize the retained random attractor and SRB measure for the discrete random dynamical system, revealing structures consistent with the theoretical chaotic phase.

math.DS↗
arXiv2025

Insights from the "Red devil" AT 2022fpx: A Dust-reddened Family of Tidal Disruption Events Excluded by Their Apparent Red Color?

We report unnoticed but intriguing features in the peculiar nuclear transient AT 2022fpx, and investigate its type. These features include the constantly red optical color of $g-r>0$, a stable soft X-ray flare ($kT\sim100$ eV) in the past $\sim$550 days, a prominent mid-infrared echo peaked at $\sim$$10^{43.3}$ erg s$^{-1}$ and the confirmation of a weak active galactic nucleus by weak flares in pre-event Wide-field Infrared Survey Explorer mid-infrared light curves with no contemporary optical, radio or X-ray counterparts. The combination of the optical red color and possible origin of a tidal disruption event (TDE) of AT 2022fpx is particularly attractive, as it challenges the most widely accepted and adopted "blue color" criterion for optical TDE selection. Although we still cannot confirm whether the red color is intrinsic, we do find that the "blue color" criterion can filter out normal TDEs whose optical-UV spectral energy distributions (SEDs) are either severely contaminated by prominent emission lines (especially H$α$) or heavily dust-reddened. Hence, its potential selection effect may have been imprinted on the whole optical TDE family. Blackbody fitting on the optical (rest-frame $\sim$$4000-7000$ Å) and optical-UV ($\sim$$2000-7000$ Å) SEDs of four TDEs with high-cadence UV observations shows that $T_\mathrm{bb}$ rise by $\sim$40$-$110 \% when the UV bands are included. The power-law models ($f_λ\proptoλ^{-α}$ with $α=2-3$) can fit the rest-frame $\sim$$2000-7000$ ÅSEDs more consistently, indicating that SEDs should peak at shorter wavelengths, but not simple blackbodies. Hence, the estimated released energy for the optical-UV bright but X-ray faint TDEs based on blackbody SED fitting should be significantly lower than the intrinsic energy.

astro-ph.HE↗
arXiv2025

What is "Spatial" about Spatial Computing?

Recent advancements in geographic information systems and mixed reality technologies have positioned spatial computing as a transformative paradigm in computational science. However, the field remains conceptually fragmented, with diverse interpretations across disciplines like Human-Computer Interaction, Geographic Information Science, and Computer Science, which hinders a comprehensive understanding of spatial computing and poses challenges for its coherent advancement and interdisciplinary integration. In this paper, we trace the origins and historical evolution of spatial computing and examine how "spatial" is understood, identifying two schools of thought: "spatial" as the contextual understanding of space, where spatial data guides interaction in the physical world; and "spatial" as a mixed space for interaction, emphasizing the seamless integration of physical and digital environments to enable embodied engagement. By synthesizing these perspectives, we propose spatial computing as a computational paradigm that redefines the interplay between environment, computation, and human experience, offering a holistic lens to enhance its conceptual clarity and inspire future technological innovations that support meaningful interactions with and shaping of environments.

cs.HC↗
arXiv2025

An efficient and robust high-order compact ALE gas-kinetic scheme for unstructured meshes

For the arbitrary-Lagrangian-Eulerian (ALE) calculations, the geometric information needs to be calculated at each time step due to the movement of mesh. To achieve the high-order spatial accuracy, a large number of matrix inversions are needed, which affect the efficiency of computation dramatically. In this paper, an efficient and robust high-order compact ALE gas-kinetic scheme is developed for the compressible moving grids and moving boundary problems. The memory-reduction reconstruction is used to construct a quadratic polynomial on the target cell, where both structured and unstructured meshes can be used. Taking derivatives of the candidate polynomial, the quadratic terms can be obtained by the least square method using the average gradient values of the cell itself and its adjacent cells. Moving the quadratic terms to right-hand side of the constrains for cell averaged value, the linear terms of the polynomial can be determined by the least square method as well. The gradient compression factor is adopted to suppress the spurious oscillations near discontinuities. Combined with the two-stage fourth-order time discretization, a high-order compact gas-kinetic scheme is developed for ALE computation. In the process of mesh movement, the inversions of lower order matrix are needed for the least square method, which makes a 7x speedup and improves the efficiency greatly. In the computation, the grid velocity can be given by the mesh adaptation method and the cell centered Lagrangian nodal solver. Numerical examples are presented to evaluate the accuracy, efficiency, robustness and the preservation of geometric conservation law of the current scheme.

physics.comp-ph↗