arXiv subjects

Yibo Wang

Publications and source records attributed to Yibo Wang.

At least 73 records · Page 4Linked to original sources

arXiv2026

WAter: A Workload-Adaptive Knob Tuning System based on Workload Compression

Selecting appropriate values for the configurable parameters of Database Management Systems (DBMS) to improve performance is a significant challenge. Recent machine learning (ML)-based tuning systems have shown strong potential, but their practical adoption is often limited by the high tuning cost. This cost arises from two main factors: (1) the system needs to evaluate a large number of configurations to identify a satisfactory one, and (2) for each configuration, the system must execute the entire target workload on the DBMS, which is both time-consuming. Existing studies have primarily addressed the first factor by improving sample efficiency, that is, by reducing the number of configurations evaluated. However, the second factor, improving runtime efficiency by reducing the time required for each evaluation, has received limited attention and remains an underexplored direction. We develop WAter, a runtime-efficient and workload-adaptive tuning system that finds near-optimal configurations at a fraction of the tuning cost compared with state-of-the-art methods. We divide the tuning process into multiple time slices and evaluate only a small subset of queries from the workload in each slice. Different subsets are evaluated across slices, and a runtime profile is used to dynamically identify more representative subsets for evaluation in subsequent slices. At the end of each time slice, the most promising configurations are evaluated on the original workload to measure their actual performance. Evaluations demonstrate that WAter identifies the best-performing configurations with up to 73.5% less tuning time and achieves up to 16.2% higher performance than the best-performing alternative.

cs.DB↗

arXiv2026

Near-Field Multiuser Beam Training for XL-MIMO: An End-to-End Interference-Aware Approach with Pilot Limitations

Near-field propagation in extremely large-scale MIMO (XL-MIMO) enlarges the beam training (BT) search space by introducing an additional range dimension, which makes conventional codebook-based beam sweeping prohibitively expensive under limited pilot resources, especially for multiuser sub-connected hybrid architectures. This letter proposes a deep-learning-based interference-aware multiuser BT framework (DL-IABT) that directly predicts analog beam indices from a small number of uplink sensing measurements. By exploiting a subarray-level approximation, a far-field codebook is adopted to represent each subarray response with negligible mismatch. To enable end-to-end (E2E) learning, we derive a variant-MSE surrogate loss by eliminating the digital precoder through a closed-form MMSE solution from KKT conditions, which implicitly accounts for multiuser interference (MUI). The proposed network integrates a complex-valued sensing front-end, a shared complex-valued encoder, a Transformer-based multiuser predictor, and a scalable Gumbel--Softmax beam selection head. Simulation results show that DL-IABT achieves near-optimal sum-rate performance while providing markedly higher effective throughput under pilot overhead constraints.

eess.SP↗

arXiv2026

MM-DeepResearch: A Simple and Effective Multimodal Agentic Search Baseline

We aim to develop a multimodal research agent capable of explicit reasoning and planning, multi-tool invocation, and cross-modal information synthesis, enabling it to conduct deep research tasks. However, we observe three main challenges in developing such agents: (1) scarcity of search-intensive multimodal QA data, (2) lack of effective search trajectories, and (3) prohibitive cost of training with online search APIs. To tackle them, we first propose Hyper-Search, a hypergraph-based QA generation method that models and connects visual and textual nodes within and across modalities, enabling to generate search-intensive multimodal QA pairs that require invoking various search tools to solve. Second, we introduce DR-TTS, which first decomposes search-involved tasks into several categories according to search tool types, and respectively optimize specialized search tool experts for each tool. It then recomposes tool experts to jointly explore search trajectories via tree search, producing trajectories that successfully solve complex tasks using various search tools. Third, we build an offline search engine supporting multiple search tools, enabling agentic reinforcement learning without using costly online search APIs. With the three designs, we develop MM-DeepResearch, a powerful multimodal deep research agent, and extensive results shows its superiority across benchmarks. Code is available at https://github.com/HJYao00/MM-DeepResearch

cs.CV↗

arXiv2026

M$^2$: Dual-Memory Augmentation for Long-Horizon Web Agents via Trajectory Summarization and Insight Retrieval

Multimodal Large Language Models (MLLMs) based agents have demonstrated remarkable potential in autonomous web navigation. However, handling long-horizon tasks remains a critical bottleneck. Prevailing strategies often rely heavily on extensive data collection and model training, yet still struggle with high computational costs and insufficient reasoning capabilities when facing complex, long-horizon scenarios. To address this, we propose M$^2$, a training-free, memory-augmented framework designed to optimize context efficiency and decision-making robustness. Our approach incorporates a dual-tier memory mechanism that synergizes Dynamic Trajectory Summarization (Internal Memory) to compress verbose interaction history into concise state updates, and Insight Retrieval Augmentation (External Memory) to guide the agent with actionable guidelines retrieved from an offline insight bank. Extensive evaluations across WebVoyager and OnlineMind2Web demonstrate that M$^2$ consistently surpasses baselines, yielding up to a 19.6% success rate increase and 58.7% token reduction for Qwen3-VL-32B, while proprietary models like Claude achieve accuracy gains up to 12.5% alongside significantly lower computational overhead.

cs.CV↗

arXiv2026

Building Autonomous GUI Navigation via Agentic-Q Estimation and Step-Wise Policy Optimization

Recent advances in Multimodal Large Language Models (MLLMs) have substantially driven the progress of autonomous agents for Graphical User Interface (GUI). Nevertheless, in real-world applications, GUI agents are often faced with non-stationary environments, leading to high computational costs for data curation and policy optimization. In this report, we introduce a novel MLLM-centered framework for GUI agents, which consists of two components: agentic-Q estimation and step-wise policy optimization. The former one aims to optimize a Q-model that can generate step-wise values to evaluate the contribution of a given action to task completion. The latter one takes step-wise samples from the state-action trajectory as inputs, and optimizes the policy via reinforcement learning with our agentic-Q model. It should be noticed that (i) all state-action trajectories are produced by the policy itself, so that the data collection costs are manageable; (ii) the policy update is decoupled from the environment, ensuring stable and efficient optimization. Empirical evaluations show that our framework endows Ovis2.5-9B with powerful GUI interaction capabilities, achieving remarkable performances on GUI navigation and grounding benchmarks and even surpassing contenders with larger scales.

cs.AI↗

arXiv2026

CFHT MegaCam Two Deep Fields Imaging Survey (2DFIS) I: Overview

We present the Two Deep Fields Imaging Survey (2DFIS), a wide-field imaging program conducted with the Canada-France-Hawaii Telescope (CFHT) targeting two astrophysically distinct regions: one containing a repeating fast radio burst (FRB) source and another hosting a candidate of a rotating galaxy cluster. Achieving a depth of r~26mag, the survey enables a search for faint optical counterparts and environmental signatures associated with the FRB, while high-quality photometric and galaxy shape measurements in the cluster field support a weak-lensing analysis of its mass distribution. This paper describes the observing strategy and data processing methodology adopted for 2DFIS, including the use of the LSST Science Pipelines with survey-specific adaptations for CFHT/MegaCam data. We outline a complete workflow for transforming raw CFHT exposures into science-ready data products, including calibrated single-epoch images, multi-band coadded mosaics, and extensive source catalogs. These data products provide the foundation for ongoing and future studies of FRB host environments, cluster mass reconstruction, and related cosmological applications.

astro-ph.GA↗

arXiv2026

Coexistence of Antiferromagnetic Spin Fluctuations and Superconductivity in La2SmNi2O7 Thin Films

The interplay between magnetic fluctuations and superconductivity is fundamental for understanding unconventional high-temperature superconductors. In the recently discovered Ruddlesden-Popper phase nickelates-which achieve superconducting transition temperatures up to ~100 K-this connection has been theoretically predicted but experimentally unverified. Using compressively strained La2SmNi2O7 thin films, we report direct evidence for this interplay. We observe a characteristic 'Mexican hat'-shaped magnetoresistance, a signature of superconductivity coexisting with emergent antiferromagnetic (AFM) fluctuations. This distinctive feature arises from a competition between the magnetic field's suppression of AFM fluctuation-driven scattering and its suppression of superconducting fluctuations. We quantify these AFM fluctuations by a crossover field, B*, whose magnitude decreases with increasing temperature and vanishes near the superconducting onset-behavior contrasting sharply with that in high-Tc cuprates. Our results establish not merely coexistence, but also a direct and novel correlation between AFM instability and superconductivity in nickelates, offering crucial insight into their pairing mechanism.

cond-mat.supr-con↗

arXiv2026

VTC-R1: Vision-Text Compression for Efficient Long-Context Reasoning

Long-context reasoning has significantly empowered large language models (LLMs) to tackle complex tasks, yet it introduces severe efficiency bottlenecks due to the computational complexity. Existing efficient approaches often rely on complex additional training or external models for compression, which limits scalability and discards critical fine-grained information. In this paper, we propose VTC-R1, a new efficient reasoning paradigm that integrates vision-text compression into the reasoning process. Instead of processing lengthy textual traces, VTC-R1 renders intermediate reasoning segments into compact images, which are iteratively fed back into vision-language models as "optical memory." We construct a training dataset based on OpenR1-Math-220K achieving 3.4x token compression and fine-tune representative VLMs-Glyph and Qwen3-VL. Extensive experiments on benchmarks such as MATH500, AIME25, AMC23 and GPQA-D demonstrate that VTC-R1 consistently outperforms standard long-context reasoning. Furthermore, our approach significantly improves inference efficiency, achieving 2.7x speedup in end-to-end latency, highlighting its potential as a scalable solution for reasoning-intensive applications. Our code is available at https://github.com/w-yibo/VTC-R1.

cs.CL↗

arXiv2026

A Tidal Disruption Event from an Intermediate-mass Black Hole Revealed by Comprehensive Multi-wavelength Observations

Tidal disruption events (TDEs) occur when a star crosses the tidal radius of a black hole (BH) and is ripped apart, providing a powerful way to probe dormant BHs over a wide mass range. In this study, we present our late-time observations and comprehensive multi-wavelength analyses of AT2018cqh, a TDE at the center of a dwarf galaxy that exhibited successive flares in the optical, X-ray, and radio bands. We discovered an unexpected high-state X-ray plateau phase following the peak until the present time. Along with its reported prolonged rise lasting at least 550 days, these unique characteristics are consistent with the scenario of a TDE caused by an intermediate-mass black hole (IMBH) with a mass of approximately $(1-6) \times 10^5$ solar masses. Furthermore, scaling relations derived from the host-galaxy properties indicated a similar BH mass in concert. This discovery highlights the invaluable role of TDEs in the search for elusive IMBHs.

astro-ph.HE↗

arXiv2026

Panacea: Mitigating Harmful Fine-tuning for Large Language Models via Post-fine-tuning Perturbation

Harmful fine-tuning attack introduces significant security risks to the fine-tuning services. Main-stream defenses aim to vaccinate the model such that the later harmful fine-tuning attack is less effective. However, our evaluation results show that such defenses are fragile--with a few fine-tuning steps, the model still can learn the harmful knowledge. To this end, we do further experiment and find that an embarrassingly simple solution--adding purely random perturbations to the fine-tuned model, can recover the model from harmful behaviors, though it leads to a degradation in the model's fine-tuning performance. To address the degradation of fine-tuning performance, we further propose Panacea, which optimizes an adaptive perturbation that will be applied to the model after fine-tuning. Panacea maintains model's safety alignment performance without compromising downstream fine-tuning performance. Comprehensive experiments are conducted on different harmful ratios, fine-tuning tasks and mainstream LLMs, where the average harmful scores are reduced by up-to 21.2%, while maintaining fine-tuning performance. As a by-product, we analyze the adaptive perturbation and show that different layers in various LLMs have distinct safety affinity, which coincide with finding from several previous study. Source code available at https://github.com/w-yibo/Panacea.

cs.CL↗

arXiv2026

DeepResearchEval: An Automated Framework for Deep Research Task Construction and Agentic Evaluation

Deep research systems are widely used for multi-step web research, analysis, and cross-source synthesis, yet their evaluation remains challenging. Existing benchmarks often require annotation-intensive task construction, rely on static evaluation dimensions, or fail to reliably verify facts when citations are missing. To bridge these gaps, we introduce DeepResearchEval, an automated framework for deep research task construction and agentic evaluation. For task construction, we propose a persona-driven pipeline generating realistic, complex research tasks anchored in diverse user profiles, applying a two-stage filter Task Qualification and Search Necessity to retain only tasks requiring multi-source evidence integration and external retrieval. For evaluation, we propose an agentic pipeline with two components: an Adaptive Point-wise Quality Evaluation that dynamically derives task-specific evaluation dimensions, criteria, and weights conditioned on each generated task, and an Active Fact-Checking that autonomously extracts and verifies report statements via web search, even when citations are missing.

cs.CL↗

arXiv2026

Triplets Better Than Pairs: Towards Stable and Effective Self-Play Fine-Tuning for LLMs

Recently, self-play fine-tuning (SPIN) has been proposed to adapt large language models to downstream applications with scarce expert-annotated data, by iteratively generating synthetic responses from the model itself. However, SPIN is designed to optimize the current reward advantages of annotated responses over synthetic responses at hand, which may gradually vanish during iterations, leading to unstable optimization. Moreover, the utilization of reference policy induces a misalignment issue between the reward formulation for training and the metric for generation. To address these limitations, we propose a novel Triplet-based Self-Play fIne-tuNing (T-SPIN) method that integrates two key designs. First, beyond current advantages, T-SPIN additionally incorporates historical advantages between iteratively generated responses and proto-synthetic responses produced by the initial policy. Even if the current advantages diminish, historical advantages remain effective, stabilizing the overall optimization. Second, T-SPIN introduces the entropy constraint into the self-play framework, which is theoretically justified to support reference-free fine-tuning, eliminating the training-generation discrepancy. Empirical results on various tasks demonstrate not only the superior performance of T-SPIN over SPIN, but also its stable evolution during iterations. Remarkably, compared to supervised fine-tuning, T-SPIN achieves comparable or even better performance with only 25% samples, highlighting its effectiveness when faced with scarce annotated data.

cs.CL↗

arXiv2026

Dust-obscured radio-emitting tidal disruption event coincident with a high-energy neutrino event

Despite the growing number of high-energy neutrinos (TeV-PeV) detected by IceCube, their astrophysical origins remain largely unidentified. Recent observations have linked a few tidal disruption events (TDEs) to the production of high-energy neutrino emission, all of which display dust-reprocessed infrared flares, indicating a dust- and gas-rich environment. By cross-matching the neutrino events and a sample of mid-infrared outbursts in nearby galaxies with transient radio flares, we uncover an optically obscured TDE candidate, SDSS J151345.75 $+$ 311125.2, which shows both spatial and temporal coincidence with the sub-PeV neutrino event IC170514B. Using a standard equipartition analysis of the synchrotron spectral evolution spanning 605 days post mid-infrared discovery, we find a little evolution in the radio-emitting region, with a kinetic energy up to $10^{51}$ erg, depending on the outflow geometry and shock acceleration efficiency assumed. High-resolution European VLBI Network imaging reveals a compact radio emission that is unresolved at a scale of $<$ 2.1 pc, with a brightness temperature of $T_b>5\times10^6$ K, suggesting that the observed late-time radio emission might originate from the interaction between a decelerating outflow and a dense circumnuclear medium. If the association is genuine, the neutrino production is possibly related to the acceleration of protons through pp collisions during the outflow expanding process, implying that the outflow-cloud interaction could provide a physical site with a high-density environment for producing the sub-PeV neutrinos. Such a scenario can be tested with future identifications of radio transients coincident with high-energy neutrinos.

astro-ph.HE↗

arXiv2026

A Study on the Triggering of Nucleonic Direct Urca Processes in Neutron Stars of Specific Masses and Their Hyperon Dependence

This work aims to analyze how hyperons affect neutrino radiation properties in nucleonic direct URCA processes, expecting to provide useful references for finding evidence of the existence of hyperons in astronomical observations. This analysis is carried out using the GM1 and NL3 parameter sets under the SU(6) and SU(3) flavor symmetries in the relativistic mean field theory framework. Combined with the inferred mass and radius values of PSRs J1231-1411, J0030+0451, and J0740+6620, our results show that nucleonic direct Urca processes are absent in PSR J1231-1411 due to momentum conservation violation. In hyperon-containing PSR J0030+0451 (NL3 parameter set), the nucleonic direct Urca processes involving $e^{-}$/ $μ^{-}$ would occur. A large inferred mass span induces hyperon fraction variations, affecting neutrino emissivity. If the inferred mass of PSR J0030+0451 exceeds approximately 1.8 $M_{\odot}$, the neutrino luminosity of the nucleonic direct Urca processes under the SU(3) flavor symmetry remains nearly the same as that in npe$μ$ matter, without depending on hyperons. However, it exhibits an obvious hyperon dependence under the SU(6) spin-flavor symmetry. For hyperon-containing J0740+6620, the nucleonic direct Urca processes under the SU(3) flavor symmetry in GM1 parameter set predicts faster neutrino luminosity decline with hyperonic fraction than npe$μ$ matter, and under the SU(6) spin-flavor symmetry in NL3 parameter set it shows monotonic decreasing trend. The research shows that hyperonic fraction significantly affect the neutrino radiation properties of the nucleonic direct URCA processes in neutron stars. Different-mass pulsars (e.g., PSRs J1231-1411, J0030+0451, J0740+6620) exhibit the distinct nucleonic direct URCA processes behaviors, dependent on inferred masses/radii, parameter sets, and theoretical models.

nucl-th↗

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↗