arXiv subjects

Jiaxing Zhao

Publications and source records attributed to Jiaxing Zhao.

At least 37 records · Page 2Linked to original sources

arXiv2026

Wan-Image: Pushing the Boundaries of Generative Visual Intelligence

We present Wan-Image, a unified visual generation system explicitly engineered to paradigm-shift image generation models from casual synthesizers into professional-grade productivity tools. While contemporary diffusion models excel at aesthetic generation, they frequently encounter critical bottlenecks in rigorous design workflows that demand absolute controllability, complex typography rendering, and strict identity preservation. To address these challenges, Wan-Image features a natively unified multi-modal architecture by synergizing the cognitive capabilities of large language models with the high-fidelity pixel synthesis of diffusion transformers, which seamlessly translates highly nuanced user intents into precise visual outputs. It is fundamentally powered by large-scale multi-modal data scaling, a systematic fine-grained annotation engine, and curated reinforcement learning data to surpass basic instruction following and unlock expert-level professional capabilities. These include ultra-long complex text rendering, hyper-diverse portrait generation, palette-guided generation, multi-subject identity preservation, coherent sequential visual generation, precise multi-modal interactive editing, native alpha-channel generation, and high-efficiency 4K synthesis. Across diverse human evaluations, Wan-Image exceeds Seedream 5.0 Lite and GPT Image 1.5 in overall performance, reaching parity with Nano Banana Pro in challenging tasks. Ultimately, Wan-Image revolutionizes visual content creation across e-commerce, entertainment, education, and personal productivity, redefining the boundaries of professional visual synthesis.

cs.CV↗

arXiv2026

LLaVA-Octopus: Unlocking Instruction-Driven Adaptive Projector Fusion for Video Understanding

In this paper, we introduce LLaVA-Octopus, a novel video multimodal large language model. LLaVA-Octopus adaptively weights features from different visual projectors based on user instructions, enabling us to leverage the complementary strengths of each projector. We observe that different visual projectors exhibit distinct characteristics when handling specific tasks. For instance, some projectors excel at capturing static details, while others are more effective at processing temporal information, and some are better suited for tasks requiring temporal coherence. By dynamically adjusting feature weights according to user instructions, LLaVA-Octopus dynamically selects and combines the most suitable features, significantly enhancing the model's performance in multimodal tasks. Experimental results demonstrate that LLaVA-Octopus achieves excellent performance across multiple benchmarks, especially in tasks such as video question answering, long video understanding, and comprehensive multi-choices benchmarks, highlighting its broad application potential.

cs.CV↗

arXiv2026

Systematic study of flow of protons and light clusters in intermediate-energy heavy-ion collisions with momentum-dependent potentials

We study the influence of the nuclear equation-of-state (EoS) on collective observables -- the directed ($v_1$) and elliptic flow ($v_2$) of nucleons and light clusters -- in heavy-ion collisions at GeV energies using the Parton-Hadron-Quantum-Molecular Dynamics (PHQMD) approach. A novel development in this work is the inclusion of a momentum-dependent nucleon potential in the PHQMD in addition to the static, density-dependent Skyrme interaction. This enables three distinct EoS scenarios: two static ("soft" and "hard", differing in compressibility) and a soft, momentum-dependent EoS calibrated to $pA$ elastic scattering data. We find a strong EoS sensitivity in proton and cluster rapidity and $p_T$ distributions: soft and soft momentum-dependent EoS yield similar results, markedly different from the hard EoS. Softening the EoS reduces proton yields at midrapidity while enhancing light-cluster production. The EoS also affects flow observables differently for nucleons and clusters. For protons, a soft momentum-dependent potential increases slightly the magnitude of $v_1$ and $v_2$ relative to the hard EoS, whereas cluster flows are nearly similar. The soft momentum-dependent EoS provides an overall good agreement with experimental data from HADES and FOPI Collaborations while the soft EOS is not in line with the data. A scaling of $v_2$ with cluster mass number $A$ is observed at midrapidity for low $p_T$, which breaks at higher $p_T$. Finally, we examine the sensitivity of flow observables to deuteron production mechanisms. Deuterons formed via MST clustering exhibit different flow patterns from those produced by coalescence at freeze-out, indicating that flow harmonics may help discriminate between cluster formation scenarios.

nucl-th↗

arXiv2026

Relativistic corrections to hadron-hadron correlation function

Femtoscopy offers a sensitive probe of hadron emission sources and hadronic interactions. In this study, we examine relativistic corrections to scattering phase shifts and correlation functions using the two-body Dirac equation framework. We analyze the impact of the Darwin term and spin-dependent potentials, showing that these relativistic effects, especially spin-related interactions, significantly enhance the proton-proton correlation function. Our findings emphasize the necessity of including relativistic corrections for precise femtoscopic analyses.

hep-ph↗

arXiv2026

Cochain: Balancing Insufficient and Excessive Collaboration in LLM Agent Workflows

Large Language Models (LLMs) have demonstrated impressive performance in executing complex reasoning tasks. Chain-of-thought effectively enhances reasoning capabilities by unlocking the potential of large models, while multi-agent systems provide more comprehensive solutions by integrating the collective intelligence of multiple agents. However, both approaches face significant limitations. Single-agent with chain-of-thought, due to the inherent complexity of designing cross-domain prompts, faces collaboration challenges. Meanwhile, multi-agent systems consume substantial tokens and inevitably dilute the primary problem, which is particularly problematic in business workflow tasks. To address these challenges, we propose Cochain, a collaboration prompting framework that effectively solves the business workflow collaboration problem by combining knowledge and prompts at a reduced cost. Specifically, we construct an integrated knowledge graph that incorporates knowledge from multiple stages. Furthermore, by maintaining and retrieving a prompts tree, we can obtain prompt information relevant to other stages of the business workflow. We perform extensive evaluations of Cochain across multiple datasets, demonstrating that Cochain outperforms all baselines in both prompt engineering and multi-agent LLMs. Additionally, expert evaluation results indicate that the use of a small model in combination with Cochain outperforms GPT-4.

cs.CL↗

arXiv2026

Rotation induced color confinement

The rotation effect on the QCD properties is an open question. We study the dynamic gluon mass in a dense QCD matter, the rotation is introduced by taking a covariant transformation between the flat and curved spaces. The law of causality which restricts the rotation strength of the system is carefully considered in the calculation. we find that the rotation effect is not monotonous. Overall, it behaves like an anti-screening effect, reflecting in the decreasing gluon mass, but the strength changes with the rotation. For a QCD matter with low baryon density, the screening effect in the flat space can be completely canceled by the rotation, and gluons are confined in a strongly rotating matter. When the rotation is extremely high, the matter approaches to a weakly interacting gas.

hep-ph↗

arXiv2026

MetaGen: Self-Evolving Roles and Topologies for Multi-Agent LLM Reasoning

Large language models are increasingly deployed as multi-agent systems, where specialized roles communicate and collaborate through structured interactions to solve complex tasks that often exceed the capacity of a single agent. However, most existing systems still rely on a fixed role library and an execution-frozen interaction topology, a rigid design choice that frequently leads to task mismatch, prevents timely adaptation when new evidence emerges during reasoning, and further inflates inference cost. We introduce MetaGen, a training-free framework that adapts both the role space and the collaboration topology at inference time, without updating base model weights. MetaGen generates and rewrites query-conditioned role specifications to maintain a controllable dynamic role pool, then instantiates a constrained execution graph around a minimal backbone. During execution, it iteratively updates role prompts and adjusts structural decisions using lightweight feedback signals. Experiments on code generation and multi-step reasoning benchmarks show that MetaGen improves the accuracy and cost tradeoff over strong multi-agent baselines.

cs.CL↗

arXiv2025

Probing of EoS with clusters and hypernuclei

The study of the nuclear equation-of-state (EoS) is a one of the primary goals of experimental and theoretical heavy-ion physics. The comparison of recent high statistics data from the STAR Collaboration with transport models provides a unique possibility to address this topic in a yet unexplored energy domain. Employing the microscopic N-body Parton-Hadron-Quantum-Molecular Dynamics (PHQMD) transport approach, which allows to describe the propagation and interactions of hadronic and partonic degrees of freedom including cluster and hyper-nucleus formation and dynamics, we investigate the influence of different EoS on bulk observables, the multiplicity, $p_T$ and rapidity distributions of protons, $Λ$s and clusters up to A=4 as well as their influence on the collective flow. We explore three different EoS: two static EoS, dubbed 'soft' and 'hard', which differ in the compressibility modulus, as well as a soft momentum dependent EoS. We find that a soft momentum dependent EoS reproduces most baryon and cluster observables, including the flow observables, quantitatively, however, hard EOS show a similar trend.

nucl-th↗

arXiv2025

Bayesian Inference of Heavy-Quark Dissipation and Jet Transport Parameters from D-Meson observables in heavy-ion collisions at the LHC energies

We perform the first simultaneous Bayesian inference of the temperature-dependent heavy-quark spatial diffusion coefficient $2πT\mathcal{D}_s$ and the scaled jet transport coefficient $\hat{q}/T^3$ in the quark-gluon plasma, utilizing $D$-meson nuclear modification factor $R_\text{AA}$ and elliptic flow $v_2$ data from Pb-Pb collisions at $\sqrt{s_\text{NN}} = 5.02\ \text{TeV}$. The analysis employs a unified improved Langevin transport model that incorporates both collisional and radiative energy loss, followed by coalescence plus fragmentation hadronization. The posterior distributions of the parameters of $\hat{q}/T^3$ and those of $2πT\mathcal{D}_s$ are well constrained, and compared with the results of theoretical models or other experimental data extraction, respectively. The $30-50\%$ centrality data provide significantly stronger constraints than the $0-10\%$ data. The extracted ratio $\hat{q}/κ$ between the quark jet transport and heavy-quark diffusion coefficients exhibits a non-monotonic temperature dependence, deviating from the value $2$ estimated from the definition, with a value interval spanning 0.25--0.8 corresponding to the mean values of the inferred parameters. This work establishes a data-driven quantitative relationship between these two fundamental transport properties in the same observables, offering crucial insight into their interplay in the strongly coupled medium.

nucl-th↗

arXiv2025

The influence of heavy quark potential on quarkonium production in quark-gluon plasma

Remler formalism is the Wigner projection of a two particle state to bound states, which is carried out when the bound state begins to exist or when one of the two particles scatters in medium. This method has been successfully applied to quarkonium production in box simulations and in heavy-ion collisions. In this study this method is extended to strongly bound states with considerable binding energies by taking into account the potential energy between heavy quark pairs in the quark-gluon plasma (QGP). We find that an attractive heavy-quark potential enhances quarkonium production and the results are consistent with those from the statistical model in box simulations, if a proper spatial cutoff is introduced to the potential to mimic quantum effects.

nucl-th↗

arXiv2025

Correlations between heavy mesons and the creation of the charmonia, bottomonia, and $B_c$ mesons in high energy $pp$ collisions

The different QCD processes, which can produce a heavy quark-antiquark ($Q\bar Q$) pair, induce different correlations between the heavy quarks. Employing the EPOS4HQ event generator we study the consequences of these correlations and compare the calculation with experimental results on open and hidden heavy flavour mesons, measured in proton-proton (pp) collisions at RHIC and LHC energies. We find that the measured correlations between heavy mesons are a direct image of the different production mechanisms, which contribute also in a different way to the transverse momentum distribution of open and hidden heavy flavour mesons. The latter are calculated in a Wigner density approach which also enables to reproduce quantitatively the measured $B_c$ spectra. This agreement allows conclusions on the spatial distribution of the heavy quark creation processes.

hep-ph↗

arXiv2025

Heavy flavor correlations and Quarkonia production in high energy pp collisions in the EPOS4 framework

In QCD, the production of heavy quark-antiquark pairs can proceed through different mechanisms, each imprinting characteristic correlations between the heavy quarks. In this work, we use the EPOS4 event generator to study how these correlations affect quarkonium production in pp collisions. Our results demonstrate that the observed correlations between heavy mesons directly reflect the underlying production mechanisms and simultaneously shape the transverse-momentum distributions of quarkonia.

hep-ph↗

arXiv2025

MSCoRe: A Benchmark for Multi-Stage Collaborative Reasoning in LLM Agents

Large Language Models (LLMs) have excelled in question-answering (QA) tasks within single domains. However, their reasoning and coordination capabilities in complex, multi-stage scenarios remain underexplored. Existing benchmarks typically focus on isolated tasks or narrow domains, overlooking models' abilities for multi-stage collaboration and optimization without explicit external guidance. To bridge this gap, we propose \textbf{MSCoRe}, a novel benchmark comprising 126696 domain-specific QA instances spanning scenarios in automotive, pharmaceutical, electronics, and energy sectors. The dataset is created using a structured three-phase pipeline: dynamic sampling, iterative question-answer generation, and a multi-level quality assessment to ensure data quality. Tasks are further categorized into three difficulty levels according to stage coverage and complexity. With MSCoRe, we have conducted a comprehensive evaluation of various state-of-the-art LLM agents. The commercial models performed best across all tasks and scenarios, but a notable gap in ROUGE scores remains between simple and complex tasks. We also tested the models' robustness and found that their performance is negatively affected by noisy data. MSCoRe provides a valuable new resource for the community to evaluate and improve multi-stage reasoning in LLM agents. The code and data are available at https://github.com/D3E0-source/MSCoRE.

cs.CL↗

arXiv2025

Wigner Phase-Space Densities of Nuclear Clusters and Hypernuclei

We solve the Schrödinger equation for few-body systems to obtain the wave function for light nuclear clusters and hypernuclei from d to $\rm ^5_{ΛΛ}He$ employing realistic nucleon-nucleon and nucleon-$Λ$ potentials. We project the solution to the hyperspherical harmonic basis states to obtain the corresponding density matrices and the Wigner densities. The experimental root mean square (rms) radii and binding energies of the different clusters are well reproduced. The Wigner densities obtained will allow to improve the present coalescence approaches to identify clusters, created in heavy-ion collisions.

nucl-th↗

arXiv2025

LLaVA-Scissor: Token Compression with Semantic Connected Components for Video LLMs

In this paper, we present LLaVA-Scissor, a training-free token compression strategy designed for video multimodal large language models. Previous methods mostly attempt to compress tokens based on attention scores, but fail to effectively capture all semantic regions and often lead to token redundancy. Differently, we propose to leverage the Semantic Connected Components (SCC) approach that assigns tokens to distinct semantic regions within the token set, ensuring comprehensive semantic coverage. The outcome is a two-step spatio-temporal token compression strategy that utilizes SCC in both spatial and temporal domains. This strategy can effectively compress tokens by representing the entire video with a set of non-overlapping semantic tokens. We conduct extensive evaluations of the token compression capabilities of LLaVA-Scissor across diverse video understanding benchmarks, including video question answering, long video understanding, and comprehensive multi-choices benchmarks. Experimental results show that the proposed LLaVA-Scissor outperforms other token compression methods, achieving superior performance in various video understanding benchmarks, particularly at low token retention ratios. Project page: https://github.com/HumanMLLM/LLaVA-Scissor.

cs.CV↗

arXiv2025

HumanOmniV2: From Understanding to Omni-Modal Reasoning with Context

With the rapid evolution of multimodal large language models, the capacity to deeply understand and interpret human intentions has emerged as a critical capability, which demands detailed and thoughtful reasoning. In recent studies, Reinforcement Learning (RL) has demonstrated potential in enhancing the reasoning capabilities of Large Language Models (LLMs). Nonetheless, the challenges associated with adapting RL to multimodal data and formats remain largely unaddressed. In this paper, we identify two issues in existing multimodal reasoning models: insufficient global context understanding and shortcut problems. Insufficient context understanding can happen when a model misinterprets multimodal context, resulting in incorrect answers. The shortcut problem occurs when the model overlooks crucial clues in multimodal inputs, directly addressing the query without considering the multimodal information. To tackle these issues, we emphasize the necessity for the model to reason with a clear understanding of the global context within multimodal inputs. This global context understanding can effectively prevent the model from overlooking key multimodal cues and ensure a thorough reasoning process. To ensure the accurate interpretation of multimodal context information, we implement a context reward judged by a large language model, alongside format and accuracy rewards. Additionally, to improve complex reasoning capability, we employ the LLM to assess the logical reward, determining whether the reasoning process successfully integrates multimodal information with logical methods. We also introduce a reasoning omni-modal benchmark, IntentBench, aimed at evaluating models in understanding complex human intentions and emotions. Our proposed method demonstrates advanced performance across multiple omni-modal benchmarks compared to other open-source omni-modal models.

cs.CV↗

arXiv2025

Searching for True Muonium in Relativistic Heavy Ion Collisions

We investigate the production of the as-yet-undetected true muonium within the quark-gluon plasma formed in relativistic heavy-ion collisions, employing a relativistic Boltzmann transport framework coupled to viscous hydrodynamic simulations. The obtained effective cross sections for central collisions are 1.23~$μb$ in AuAu collisions with $\sqrt{s_{\rm NN}}=200$~GeV and 14.2~$μb$ in PbPb collisions with $\sqrt{s_{\rm NN}}=5.02$~TeV, resulting in a yield of $\mathcal{O}(10^4)$ and $\mathcal{O}(10^5)$ true muonium per billion $AA$ collisions at RHIC and the LHC, respectively. This establishes heavy-ion collisions as a promising process for detecting true muonium.

nucl-th↗

arXiv2025

Suppressed Drell-Yan process by an external magnetic field

The strongest electromagnetic fields in nature are created in high energy nuclear collisions and expected to change the dynamic scattering processes in the early stage. The magnetic field effect on the Drell-Yan process is investigated in this work. The single photon decay into quark pairs and lepton pairs in an external magnetic field leads to a significant Drell-Yan suppression in low and intermediate invariant mass region. The calculation up to the Landau level $n-1$ is complete in the energy region $s<2/3n(eB)$, and the enlarged phase space at higher landau levels may enhance the dilepton spectrum in the high mass region.

hep-ph↗