arXiv subjects

Ming Sun

Publications and source records attributed to Ming Sun.

At least 37 records · Page 2Linked to original sources

arXiv2026

SEPS: Semantic-enhanced Patch Slimming Framework for fine-grained cross-modal alignment

Fine-grained cross-modal alignment aims to establish precise local correspondences between vision and language, forming a cornerstone for visual question answering and related multimodal applications. Current approaches face challenges in addressing patch redundancy and ambiguity, which arise from the inherent information density disparities across modalities. Recently, Multimodal Large Language Models (MLLMs) have emerged as promising solutions to bridge this gap through their robust semantic generation capabilities. However, the dense textual outputs from MLLMs may introduce conflicts with the original sparse captions. Furthermore, accurately quantifying semantic relevance between rich visual patches and concise textual descriptions remains a core challenge. To overcome these limitations, we introduce the Semantic-Enhanced Patch Slimming (SEPS) framework, which systematically addresses patch redundancy and ambiguity. Our approach employs a two-stage mechanism to integrate unified semantics from both dense and sparse texts, enabling the identification of salient visual patches. Additionally, it leverages relevance-aware selection with mean value computation to highlight crucial patch-word correspondences, thereby improving cross-modal similarity assessment. Comprehensive experiments on Flickr30K and MS-COCO datasets validate that SEPS achieves superior performance, surpassing existing approaches by 23\%-86\% in rSum across diverse model architectures, with notable enhancements in text-to-image retrieval scenarios. Our implementation is available at https://github.com/Sweet4tars/seps.git.

cs.CV↗

arXiv2026

A merger shock traced by radio arcs and ultra-long radio tails in galaxy cluster A2142

Abell 2142 (A2142) is a massive, nearby galaxy cluster undergoing a complex merger. It exhibits an elongated X-ray morphology along the northwest-southeast axis and hosts four known cold fronts. Using XMM-Newton observations, we detect a merger shock on the northwest side of the cluster with a Mach number of $M \sim 1.3$. The observed shock front and four cold fronts can be reproduced by numerical simulations of an off-axis merger with a large impact parameter, which imparts significant angular momentum to induce the sloshing of the subcluster core and large-scale ambient gas. In projection, the shock front is spatially coincident with arc-shaped radio filaments observed behind the prominent head-tail radio galaxies T1 and T2. We interpret these radio arcs as partial vortex ring structures (resembling ``smoke rings'') produced by the interaction of the merger shock with the low-density cocoons of radio galaxies. The shock strips and rolls the jet cocoon into a toroidal vortex, as predicted by recent magnetohydrodynamic simulations. We further demonstrate that the merger shock can significantly elongate the radio tails by re-accelerating aged relativistic electrons and stretching the tail plasma via the post-shock wind. This process provides a natural explanation for the $>$500 kpc tail observed in this and other merging clusters. Our findings establish radio arcs and ultra-long radio tails as independent, complementary tracers of merger shocks in galaxy clusters. Our results demonstrate that merger shocks can reshape both the thermal and non-thermal components of galaxy clusters, and that tailed radio galaxies serve as sensitive probes of intracluster medium weather.

astro-ph.GA↗

arXiv2026

NTIRE 2024 Challenge on Image Super-Resolution (x4): Methods and Results

This paper reviews the NTIRE 2024 challenge on image super-resolution ($\times$4), highlighting the solutions proposed and the outcomes obtained. The challenge involves generating corresponding high-resolution (HR) images, magnified by a factor of four, from low-resolution (LR) inputs using prior information. The LR images originate from bicubic downsampling degradation. The aim of the challenge is to obtain designs/solutions with the most advanced SR performance, with no constraints on computational resources (e.g., model size and FLOPs) or training data. The track of this challenge assesses performance with the PSNR metric on the DIV2K testing dataset. The competition attracted 199 registrants, with 20 teams submitting valid entries. This collective endeavour not only pushes the boundaries of performance in single-image SR but also offers a comprehensive overview of current trends in this field.

cs.CV↗

arXiv2026

Accelerated Decentralized Stochastic Gradient Descent for Strongly Convex Optimization

Decentralized stochastic optimization is a fundamental paradigm for large-scale learning over networks, where agents communicate only with their neighbors and no central coordinator is required. For strongly convex problems, communication efficiency is mainly determined by the condition number $κ=L/μ$ and the network spectral gap $1-β$. Although deterministic decentralized methods can simultaneously achieve accelerated $\sqrtκ$ and $1/\sqrt{1-β}$ dependences, no existing stochastic method attains both improvements at once. In this paper, we propose \emph{Multi-Gossip Accelerated DSGD} (MG-ADSGD), a decentralized stochastic algorithm that combines Nesterov-type primal--dual extrapolation with multi-round fast gossip averaging. The key idea is to couple the gossip depth with the mini-batch size so that additional communication rounds simultaneously improve consensus accuracy and reduce gradient variance. We show that MG-ADSGD achieves the communication complexity \[ \widetilde{\mathcal O}\!\left( \frac{σ^2}{μnε}\log\frac{1}ε + \sqrt{\fracκ{1-β}}\log\frac{1}ε \right), \] where $ε$ denotes the target accuracy, $n$ is the number of nodes, and $σ^2$ is the gradient variance. To the best of our knowledge, this bound yields the best currently available communication complexity for decentralized stochastic strongly convex optimization, up to logarithmic factors that are independent of $ε$.

cs.LG↗

arXiv2026

JWST reveals how black holes are fed: kiloparsec-scale multiphase filaments feed sub-kiloparsec circumnuclear disks

The Centaurus cluster is one of the most important archetypes of radio-mode AGN feedback, with its central galaxy, NGC 4696, launching powerful jets that inflate X-ray cavities and regulate cooling and star formation. NGC 4696 lies within a spectacular multiphase nebula of filaments extending over tens of kiloparsecs and spanning six decades in temperature, from hot (10^8 K) X-ray-emitting plasma to cold molecular gas. Owing to its proximity, Hubble Space Telescope H_alpha imaging reveals a striking S-shaped ionized-gas swirl within the black hole's sphere of influence - the first such structure identified in a cluster core. Here we present the first JWST observations of NGC 4696 with NIRSpec, probing the inner 618 pc * 618 pc at 10 pc resolution. These data reveal that the ionized swirl is a rotating, multiphase circumnuclear disk (CND) physically and kinematically connected to the larger-scale filamentary network. This provides the long-sought missing link between kiloparsec-scale cooling flows and black hole accretion on <100 pc scales. Strikingly, the observed morphology and kinematics are reproduced by tailored magnetohydrodynamic simulations, in which filamentary gas condenses from the hot atmosphere, loses angular momentum, and feeds a rotating CND that mediates accretion onto the black hole. A similar structure in NGC 1275, the Perseus cluster's central galaxy, together with our results on NGC 4696 - two prototypical radio-mode AGN feedback systems - points to a common mechanism: multiphase filaments transport gas from cluster scales down to the vicinity of the black hole via a CND, closing the AGN feedback loop and establishing a physically grounded framework for self-regulated galaxy evolution.

astro-ph.GA↗

arXiv2026

MMG2Skill: Can Agents Distill In-the-Wild Guides into Self-Evolving Skills?

Abundant procedural knowledge on the Web holds great potential for helping agents solve long-horizon tasks. However, such knowledge is often multimodal, heterogeneous, noisy, and implicitly assumes human executors, making it difficult to use directly as the skills required by agents. To bridge the gap between human-oriented guides and agent-executable skills, we formalize this problem as guide-to-skill learning: converting in-the-wild guides into executable skills and continuously improving them from trajectories observable to the agent. To evaluate the capability of existing agents on this task, we introduce MMG2Skill-Bench, the first benchmark designed for this problem. We further propose MMG2Skill, a closed-loop framework that compiles guides into editable skills, conditions a fixed vision-language model (VLM) agent on these skills during execution, and revises the skills from trajectory-level root-cause feedback without using benchmark scores. Across GUI control, open-ended gameplay, and strategic card play with six VLM backbones, MMG2Skill consistently outperforms vanilla baseline agents in every model-domain setting, achieving macro-average gains of +12.8 to +25.3 percentage points across backbones. Ablation studies show that directly prompting agents with raw guides can degrade performance, while both structured skill construction and trajectory-driven revision are necessary for the observed improvements. On success-inferable tasks, analyzer-based early stopping further prevents late-stage performance regressions and saves 25%-53% of attempts when the success signal is properly calibrated.

cs.CL↗

arXiv2026

A Post-starburst Galaxy Undergoing Ram-pressure Stripping at Redshift 3.06

Understanding how galaxies ignite and extinguish their star formation remains a cornerstone question in modern astrophysics. Recent JWST surveys have revealed an overabundance of massive quiescent galaxies in the first billion years of the Universe, challenging current models of galaxy evolution. In the nearby Universe, ram pressure stripping (RPS) is a major environmental mechanism capable of rapidly shutting down star formation, yet direct observation remains scarce at redshift $z\gtrsim1$, and its role at $z>2$ is even poorly constrained by simulations. Here, we utilize JWST and ALMA observations to present direct evidence of RPS in the post-starburst galaxy A2744-JF-z3, residing in a galaxy group at redshift 3.06, the earliest such detection to date. Spectroscopic diagnostics and spectral energy distribution modeling reveal the ongoing removal of cold gas and dust, coincident with the abrupt cessation of star formation. Contrary to hydrodynamical simulations that predict a reduced incidence of RPS at high redshift, our results instead imply that RPS can operate at $z>3$, suggesting a highly stochastic and impulsive stripping within a clumpy, filamentary intra-group and circumgalactic medium. These observations extend environmental quenching well into the epoch of galaxy assembly, highlighting RPS as a previously overlooked decisive pathway to rapid quenching in nascent groups and protoclusters in the early Universe.

astro-ph.GA↗

arXiv2026

Solving the cooling flow problem with combined jet-wind AGN feedback

Active galactic nucleus (AGN) feedback is widely viewed as the most promising solution to the long-standing cooling flow problem in galaxy clusters, yet previous models prescribe jet properties inconsistent with accretion physics. We perform an idealized hydrodynamic simulation of a galaxy cluster with no merger history and a relaxed state, with its other properties similar to the Perseus cluster using the MACER framework, incorporating both jets and winds whose properties are constrained by general relativistic magnetohydrodynamic simulations of black hole accretion and observations. The combined feedback reproduces key observables, including cold gas mass, star formation rate, thermodynamic radial profiles, and black hole growth, while jet-only or wind-only models fail. The success arises from turbulence driven by jet-wind shear that enhances kinetic-to-thermal energy conversion, boosting heating efficiency by factors of three and six relative to wind-only and jet-only cases, respectively.

astro-ph.GA↗

arXiv2026

Structure-Centric Graph Foundation Model via Geometric Bases

Graph foundation models (GFMs) seek transferable representations across graph domains but are limited by structural heterogeneity and incompatible node feature spaces. We propose Structure-Centric Graph Foundation Models (SCGFM), which treat graph topology as the primary source of transferable knowledge. Modeling graphs as metric measure spaces, SCGFM introduces learnable geometric bases that define a shared structural coordinate system. Graphs are aligned to these bases via Gromov-Wasserstein distances, yielding structure-aligned latent representations that accommodate heterogeneous graph topologies. To address feature incompatibility, SCGFM employs a structure-aware feature re-encoding mechanism that unifies node representations without assuming a fixed feature dimensionality or requiring dataset-specific preprocessing. Experiments on graph- and node-level tasks demonstrate strong in-domain and cross-domain generalization, outperforming existing GFM approaches.

cs.LG↗

arXiv2026

Towards Backdoor-Based Ownership Verification for Vision-Language-Action Models

Vision-Language-Action models (VLAs) support generalist robotic control by enabling end-to-end decision policies directly from multi-modal inputs. As trained VLAs are increasingly shared and adapted, protecting model ownership becomes essential for secure deployment and responsible open-source usage. In this paper, we present GuardVLA, the first backdoor-based ownership verification framework specifically designed for VLAs. GuardVLA embeds a stealthy and harmless backdoor watermark into the protected model during training by injecting secret messages into embodied visual data. For post-release verification, we propose a swap-and-detect mechanism, in which the trigger projector and an external classifier head are used to activate and detect the embedded backdoor based on prediction probabilities. Extensive experiments across multiple datasets, model architectures, and adaptation settings demonstrate that GuardVLA enables reliable ownership verification while preserving benign task performance. Further results show that the embedded watermark remains detectable under post-release model adaptation.

cs.RO↗

arXiv2026

WebCompass: Towards Multimodal Web Coding Evaluation for Code Language Models

Large language models are rapidly evolving into interactive coding agents capable of end-to-end web coding, yet existing benchmarks evaluate only narrow slices of this capability, typically text-conditioned generation with static-correctness metrics, leaving visual fidelity, interaction quality, and codebase-level reasoning largely unmeasured. We introduce WebCompass, a multimodal benchmark that provides unified lifecycle evaluation of web engineering capability. Recognizing that real-world web coding is an iterative cycle of generation, editing, and repair, WebCompass spans three input modalities (text, image, video) and three task types (generation, editing, repair), yielding seven task categories that mirror professional workflows. Through a multi-stage, human-in-the-loop pipeline, we curate instances covering 15 generation domains, 16 editing operation types, and 11 repair defect types, each annotated at Easy/Medium/Hard levels. For evaluation, we adopt a checklist-guided LLM-as-a-Judge protocol for editing and repair, and propose a novel Agent-as-a-Judge paradigm for generation that autonomously executes generated websites in a real browser, explores interactive behaviors via the Model Context Protocol (MCP), and iteratively synthesizes targeted test cases, closely approximating human acceptance testing. We evaluate representative closed-source and open-source models and observe that: (1) closed-source models remain substantially stronger and more balanced; (2) editing and repair exhibit distinct difficulty profiles, with repair preserving interactivity better but remaining execution-challenging; (3) aesthetics is the most persistent bottleneck, especially for open-source models; and (4) framework choice materially affects outcomes, with Vue consistently challenging while React and Vanilla/HTML perform more strongly depending on task type.

cs.SE↗

arXiv2026

CodeTracer: Towards Traceable Agent States

Code agents are advancing rapidly, but debugging them is becoming increasingly difficult. As frameworks orchestrate parallel tool calls and multi-stage workflows over complex tasks, making the agent's state transitions and error propagation hard to observe. In these runs, an early misstep can trap the agent in unproductive loops or even cascade into fundamental errors, forming hidden error chains that make it hard to tell when the agent goes off track and why. Existing agent tracing analyses either focus on simple interaction or rely on small-scale manual inspection, which limits their scalability and usefulness for real coding workflows. We present CodeTracer, a tracing architecture that parses heterogeneous run artifacts through evolving extractors, reconstructs the full state transition history as a hierarchical trace tree with persistent memory, and performs failure onset localization to pinpoint the failure origin and its downstream chain. To enable systematic evaluation, we construct CodeTraceBench from a large collection of executed trajectories generated by four widely used code agent frameworks on diverse code tasks (e.g., bug fixing, refactoring, and terminal interaction), with supervision at both the stage and step levels for failure localization. Experiments show that CodeTracer substantially outperforms direct prompting and lightweight baselines, and that replaying its diagnostic signals consistently recovers originally failed runs under matched budgets. Our code and data are publicly available.

cs.SE↗

arXiv2026

KAT-Coder-V2 Technical Report

We present KAT-Coder-V2, an agentic coding model developed by the KwaiKAT team at Kuaishou. KAT-Coder-V2 adopts a "Specialize-then-Unify" paradigm that decomposes agentic coding into five expert domains - SWE, WebCoding, Terminal, WebSearch, and General - each undergoing independent supervised fine-tuning and reinforcement learning, before being consolidated into a single model via on-policy distillation. We develop KwaiEnv, a modular infrastructure sustaining tens of thousands of concurrent sandbox instances, and scale RL training along task complexity, intent alignment, and scaffold generalization. We further propose MCLA for stabilizing MoE RL training and Tree Training for eliminating redundant computation over tree-structured trajectories with up to 6.2x speedup. KAT-Coder-V2 achieves 79.6% on SWE-bench Verified (vs. Claude Opus 4.6 at 80.8%), 88.7 on PinchBench (surpassing GLM-5 and MiniMax M2.7), ranks first across all three frontend aesthetics scenarios, and maintains strong generalist scores on Terminal-Bench Hard (46.8) and tau^2-Bench (93.9). Our model is publicly available at https://streamlake.com/product/kat-coder.

cs.CL↗

arXiv2026

QPT V2: Masked Image Modeling Advances Visual Scoring

Quality assessment and aesthetics assessment aim to evaluate the perceived quality and aesthetics of visual content. Current learning-based methods suffer greatly from the scarcity of labeled data and usually perform sub-optimally in terms of generalization. Although masked image modeling (MIM) has achieved noteworthy advancements across various high-level tasks (e.g., classification, detection etc.). In this work, we take on a novel perspective to investigate its capabilities in terms of quality- and aesthetics-awareness. To this end, we propose Quality- and aesthetics-aware pretraining (QPT V2), the first pretraining framework based on MIM that offers a unified solution to quality and aesthetics assessment. To perceive the high-level semantics and fine-grained details, pretraining data is curated. To comprehensively encompass quality- and aesthetics-related factors, degradation is introduced. To capture multi-scale quality and aesthetic information, model structure is modified. Extensive experimental results on 11 downstream benchmarks clearly show the superior performance of QPT V2 in comparison with current state-of-the-art approaches and other pretraining paradigms.

cs.CV↗

arXiv2026

Score2Instruct: Scaling Up Video Quality-Centric Instructions via Automated Dimension Scoring

Classical video quality assessment methods generate a numerical score to judge a video's perceived visual fidelity and clarity. Yet, a score fails to describe the video's complex quality dimensions, restricting its applicability. Benefiting from the human-friendly linguistic output, adapting video large multimodal models to VQA via instruction tuning has the potential to address this issue. The core of the approach lies in the video quality-centric instruction data. Previous explorations mainly focus on the image domain, and their data generation processes heavily rely on human quality annotations and proprietary systems, limiting data scalability and effectiveness. To address these challenges, we propose the Score-based Instruction Generation pipeline. Specifically, SIG first scores multiple quality dimensions of an unlabeled video and maps scores to text-defined levels. It then explicitly incorporates a hierarchical Chain-of-Thought to model the correlation between specific dimensions and overall quality, mimicking the human visual system's reasoning process. The automated pipeline eliminates the reliance on expert-written quality descriptions and proprietary systems, ensuring data scalability and generation efficiency. To this end, the resulting Score2Instruct dataset contains over 320K diverse instruction-response pairs, laying the basis for instruction tuning. Moreover, to advance video LMMs' quality scoring and justification abilities simultaneously, we devise a progressive tuning strategy to fully unleash the power of S2I. Built upon SIG, we further curate a benchmark termed S2I-Bench with 400 open-ended questions to better evaluate the quality justification capacity of video LMMs. Experimental results on the S2I-Bench and existing benchmarks indicate that our method consistently improves quality scoring and justification capabilities across multiple video LMMs.

cs.CV↗

arXiv2026

Versatile Recompression-Aware Perceptual Image Super-Resolution

Perceptual image super-resolution (SR) methods restore degraded images and produce sharp outputs. In practice, those outputs are usually recompressed for storage and transmission. Ignoring recompression is suboptimal as the downstream codec might add additional artifacts to restored images. However, jointly optimizing SR and recompression is challenging, as the codecs are not differentiable and vary in configuration. In this paper, we present \textbf{Versatile Recompression-Aware Perceptual Super-Resolution (VRPSR)}, which makes existing perceptual SR aware of versatile compression. First, we formulate compression as conditional text-to-image generation and utilize a pre-trained diffusion model to build a generalizable codec simulator. Next, we propose a set of training techniques tailored for perceptual SR, including optimizing the simulator using perceptual targets and adopting slightly compressed images as the training target. Empirically, our VRPSR achieves 10% - 40% bitrate savings based on Real-ESRGAN and S3Diff under H.264/H.265/H.266 single-picture (intra) compression. Besides, our VRPSR facilitates joint optimization of SR and the post-processing model after recompression.

cs.CV↗

arXiv2026

Pioneering Perceptual Video Fluency Assessment: A Novel Task with Benchmark Dataset and Baseline

Accurately estimating humans' subjective feedback on video fluency, e.g., motion consistency and frame continuity, is crucial for various applications like streaming and gaming. Yet, it has long been overlooked, as prior arts have focused on solving it in the video quality assessment (VQA) task, merely as a sub-dimension of overall quality. In this work, we conduct pilot experiments and reveal that current VQA predictions largely underrepresent fluency, thereby limiting their applicability. To this end, we pioneer Video Fluency Assessment (VFA) as a standalone perceptual task focused on the temporal dimension. To advance VFA research, 1) we construct a fluency-oriented dataset, FluVid, comprising 4,606 in-the-wild videos with balanced fluency distribution, featuring the first-ever scoring criteria and human study for VFA. 2) We develop a large-scale benchmark of 23 methods, the most comprehensive one thus far on FluVid, gathering insights for VFA-tailored model designs. 3) We propose a baseline model called FluNet, which deploys temporal permuted self-attention (T-PSA) to enrich input fluency information and enhance long-range inter-frame interactions. Our work not only achieves state-of-the-art performance but, more importantly, offers the community a roadmap to explore solutions for VFA.

cs.CV↗

arXiv2026

HST view of NGC 5044: Constraints on Filament Widths, Magnetic Support, Multiphase Structure, and Comparison with Cluster Environments

We present new Hubble Space Telescope (HST) imaging of ionised filaments in the brightest group galaxy NGC 5044. These filaments extend several kiloparsecs and have widths of $\sim$50--120 pc, with some as narrow as those in cluster cores and others broader, reflecting the lower confining pressure in groups. Filament width ($W$) scales with ambient pressure ($P$) as $W \propto P^{-0.4}$. Combining HST, ALMA, and MUSE data, we measure column densities and magnetic field strengths. Equipartition fields decline from $\sim$40 $μ$G at the centre to $\sim$20 $μ$G at 5 kpc, about 2--3 times weaker than in clusters. Dynamical stability requires stronger radial fields ($\sim$10$^2$ $μ$G), consistent with simulations and magnetic draping, though such high values exceed Faraday Rotation Measure limits. Turbulence and cosmic rays also contribute support. Group and cluster filaments are stable against gravitational collapse, and ultraviolet imaging reveals no star formation in NGC 5044 ($<$10$^{-3}$ M$_\odot$ yr$^{-1}$). NGC 5044 hosts an ionised gas core within its Bondi radius with $n_e \propto r^{-1}$ and filling factor $f \gtrsim 3 \times 10^{-3}$, that is connected to the extended filaments, suggesting a channel for gas inflow toward the black hole. Group and cluster filaments likely share a common origin, with magnetic fields and AGN feedback preserving their structure. Ambient pressure and dust survival regulate molecular gas formation. Lower-pressure groups favour broader, more diffuse filaments with sporadic molecular clumps and weaker dust shielding, whereas higher-pressure clusters host narrower strands with stronger molecular-ionised gas alignment. We predict that (i) filament width scales with ambient pressure, (ii) filament-coincident Faraday rotation structures emerge at $\leq 0.1$ kpc resolution, and (iii) molecular/ionised gas co-spatiality is weaker in groups than in clusters.

astro-ph.GA↗