arXiv subjects

Liang Tang

Publications and source records attributed to Liang Tang.

At least 19 recordsLinked to original sources

arXiv2026

QCD Sum Rule Analysis of Triply Heavy $(Q\bar{Q})(Q\bar{q})$ Tetraquark States with $J^P=0^{\pm}$

Within the framework of QCD sum rules, we systematically investigate the mass spectra and possible decay patterns of the $(c\bar{c})(c\bar{q})$ and $(b\bar{b})(b\bar{q})$ tetraquark states with quantum numbers $J^{P}=0^{\pm}$. Based on two distinct color configurations, $[8_c]_{Q\bar{Q}} \otimes [8_c]_{Q\bar{q}}$ and $[1_c]_{Q\bar{Q}} \otimes [1_c]_{Q\bar{q}}$, we construct 18 interpolating currents for these states, and obtain stable sum rules for a subset of them. By calculating the corresponding two-point correlation functions, we extract their mass spectra. For the $(c\bar{c})(c\bar{q})$ system, we identify four possible tetraquark states: two with $J^{P}=0^+$, namely $T_{3c,0}(4760)$ and $T_{3c,0}(5000)$, and two with $J^{P}=0^-$, denoted as $T_{3c,0}(5040)$ and $T_{3c,0}(5370)$. For the $(b\bar{b})(b\bar{q})$ system, the extracted masses are found to lie in the ranges $13.72$--$14.02$ GeV for the $J^{P}=0^+$ states and $13.90$--$14.22$ GeV for the $J^{P}=0^-$ states. We further analyze their possible decay modes. Our results indicate that $T_{3c,0}(5000)$, $T_{3c,0}(5040)$, and $T_{3c,0}(5370)$ can decay into a charmonium state and a charmed meson, and are therefore expected to have appreciable decay widths. By contrast, $T_{3c,0}(4760)$ and all predicted $(b\bar{b})(b\bar{q})$ tetraquark states are expected to be relatively narrow, since the corresponding two-body strong decays via the fall-apart mechanism are kinematically forbidden. Therefore, $T_{3c,0}(4760)$ and all predicted $(b\bar{b})(b\bar{q})$ tetraquark states are promising candidates for experimental searches in final states containing a $D$ or a $\bar{B}$ meson, accompanied by light hadrons or a photon.

hep-ph↗

arXiv2026

A Turbo-Inference Strategy for Object Detection and Instance Segmentation

Object detection and instance segmentation tasks are closely related. Existing top-down instance segmentation methods usually follow a detect-then-segment paradigm, where an initial detector is used to recognize and localize objects with bounding boxes, followed by the segmentation of an instance mask within each bounding box. In such methods, the detection accuracy directly influences the subsequent segmentation performance. However, previous research has seldom explored the impact of the instance segmentation task on object detection. In this paper, we present a turbo-inference strategy for the top-down methods that leverages the complementary information between detection and segmentation tasks iteratively. Specifically we design two modules: turbo-detection head and turbo-segmentation head, which facilitate communication between the tasks. The two modules form a closed loop that interlaces the detection and segmentation results without retraining the model. Comprehensive experiments on the COCO, iFLYTEK, and Cityscapes datasets demonstrate that our method substantially enhances both detection and segmentation accuracies with a certain increase in computational cost. The proposed method represents a tradeoff between prediction accuracy and inference speed. Codes are available at https://github.com/zhaozhen2333/Turbo-Learning.git.

cs.CV↗

arXiv2026

QCD Sum Rule Analysis of a Compact $D^{+}D^{-}K^{+}$-Like Hidden-Charm Hexaquark with $J^{P}=0^{-}$

In this work, we study a compact hexaquark configuration motivated by the same quark content as $D^{+} D^{-} K^+$ using QCD sum rules, where the $D D K$ system has been extensively studied within theoretical frameworks of few-body hadronic dynamics and coupled-channel interactions. The state is constructed from three color-octet quark--antiquark clusters coupled to an overall color singlet. We construct six independent local interpolating currents with the quantum numbers $J^{P}=0^{-}$ and analyze the corresponding two-point correlation functions. Both perturbative contributions and nonperturbative condensates up to dimension ten are included in the operator product expansion. Our analysis indicates that, within Borel windows satisfying standard sum rule criteria and with a reasonable choice of continuum threshold, the mass of the $J^{P}=0^{-}$ hidden-charm hexaquark state is estimated to be in the range $3.94$--$4.41~\mathrm{GeV}$. This prediction can provide a valuable theoretical reference for identifying such exotic hidden-charm hexaquark state in future experiments.

hep-ph↗

arXiv2026

BAMI: Training-Free Bias Mitigation in GUI Grounding

GUI grounding is a critical capability for enabling GUI agents to execute tasks such as clicking and dragging. However, in complex scenarios like the ScreenSpot-Pro benchmark, existing models often suffer from suboptimal performance. Utilizing the proposed \textbf{Masked Prediction Distribution (MPD)} attribution method, we identify that the primary sources of errors are twofold: high image resolution (leading to precision bias) and intricate interface elements (resulting in ambiguity bias). To address these challenges, we introduce \textbf{Bias-Aware Manipulation Inference (BAMI)}, which incorporates two key manipulations, coarse-to-fine focus and candidate selection, to effectively mitigate these biases. Our extensive experimental results demonstrate that BAMI significantly enhances the accuracy of various GUI grounding models in a training-free setting. For instance, applying our method to the TianXi-Action-7B model boosts its accuracy on the ScreenSpot-Pro benchmark from 51.9\% to 57.8\%. Furthermore, ablation studies confirm the robustness of the BAMI approach across diverse parameter configurations, highlighting its stability and effectiveness. Code is available at https://github.com/Neur-IO/BAMI.

cs.CV↗

arXiv2026

AdaZoom-GUI: Adaptive Zoom-based GUI Grounding with Instruction Refinement

GUI grounding is a critical capability for vision-language models (VLMs) that enables automated interaction with graphical user interfaces by locating target elements from natural language instructions. However, grounding on GUI screenshots remains challenging due to high-resolution images, small UI elements, and ambiguous user instructions. In this work, we propose AdaZoom-GUI, an adaptive zoom-based GUI grounding framework that improves both localization accuracy and instruction understanding. Our approach introduces an instruction refinement module that rewrites natural language commands into explicit and detailed descriptions, allowing the grounding model to focus on precise element localization. In addition, we design a conditional zoom-in strategy that selectively performs a second-stage inference on predicted small elements, improving localization accuracy while avoiding unnecessary computation and context loss on simpler cases. To support this framework, we construct a high-quality GUI grounding dataset and train the grounding model using Group Relative Policy Optimization (GRPO), enabling the model to predict both click coordinates and element bounding boxes. Experiments on public benchmarks demonstrate that our method achieves state-of-the-art performance among models with comparable or even larger parameter sizes, highlighting its effectiveness for high-resolution GUI understanding and practical GUI agent deployment.

cs.CV↗

arXiv2026

Nuanced Emotion Recognition Based on a Segment-based MLLM Framework Leveraging Qwen3-Omni for AH Detection

Emotion recognition in videos is a pivotal task in affective computing, where identifying subtle psychological states such as Ambivalence and Hesitancy holds significant value for behavioral intervention and digital health. Ambivalence and Hesitancy states often manifest through cross-modal inconsistencies such as discrepancies between facial expressions, vocal tones, and textual semantics, posing a substantial challenge for automated recognition. This paper proposes a recognition framework that integrates temporal segment modeling with Multimodal Large Language Models. To address computational efficiency and token constraints in long video processing, we employ a segment-based strategy, partitioning videos into short clips with a maximum duration of 5 seconds. We leverage the Qwen3-Omni-30B-A3B model, fine-tuned on the BAH dataset using LoRA and full-parameter strategies via the MS-Swift framework, enabling the model to synergistically analyze visual and auditory signals. Experimental results demonstrate that the proposed method achieves an accuracy of 85.1% on the test set, significantly outperforming existing benchmarks and validating the superior capability of Multimodal Large Language Models in capturing complex and nuanced emotional conflicts. The code is released at https://github.com/dlnn123/A-H-Detection-with-Qwen-Omni.git.

cs.CV↗

arXiv2025

QCD sum rule predictions on gluonic tetraquark states with $J^{PC}=0^{+-},0^{--}$ and $1^{\pm \pm}$

In this work, we present a systematic calculation of the mass spectrum for tetraquark hybrid states, focusing on the $8_{[c\bar{c}]}\otimes 8_{[G]}\otimes 8_{[c\bar{c}]}$ color configuration, within the framework of QCD sum rules. As an extension of our previous work on $0^{++}$ and $0^{-+}$ states, we now construct 18 distinct interpolating currents with $J^{PC} = 0^{+-}$, $0^{--}$, and $1^{\pm\pm}$. Using operator product expansion (OPE) techniques and including nonperturbative contributions up to dimension six, we obtain key results: for the $0^{+-}$, $1^{--}$, and $1^{-+}$ states, the predicted masses lie in the range of $7.2-7.3$ GeV, while the $1^{+-}$ and $1^{++}$ states have slightly lower masses, between 6.9 and 7.1 GeV. These predictions provide strong support for the possible existence of an $8_{[c\bar{c}]}\otimes 8_{[G]}\otimes 8_{[c\bar{c}]}$ component within the di-$J/\psi$ structure reported by LHCb. Moreover, our analogous calculations for tetrabottom hybrid states yield mass ranges of $19.4-19.5$ GeV (for $0^{+-}$, $1^{--}$, and $1^{-+}$) and $19.2-19.3$ GeV (for $1^{+-}$ and $1^{++}$), offering crucial references for future searches.

hep-ph↗

arXiv2025

DiTraj: training-free trajectory control for video diffusion transformer

Diffusion Transformers (DiT)-based video generation models with 3D full attention exhibit strong generative capabilities. Trajectory control represents a user-friendly task in the field of controllable video generation. However, existing methods either require substantial training resources or are specifically designed for U-Net, do not take advantage of the superior performance of DiT. To address these issues, we propose DiTraj, a simple but effective training-free framework for trajectory control in text-to-video generation, tailored for DiT. Specifically, first, to inject the object's trajectory, we propose foreground-background separation guidance: we use the Large Language Model (LLM) to convert user-provided prompts into foreground and background prompts, which respectively guide the generation of foreground and background regions in the video. Then, we analyze 3D full attention and explore the tight correlation between inter-token attention scores and position embedding. Based on this, we propose inter-frame Spatial-Temporal Decoupled 3D-RoPE (STD-RoPE). By modifying only foreground tokens' position embedding, STD-RoPE eliminates their cross-frame spatial discrepancies, strengthening cross-frame attention among them and thus enhancing trajectory control. Additionally, we achieve 3D-aware trajectory control by regulating the density of position embedding. Extensive experiments demonstrate that our method outperforms previous methods in both video quality and trajectory controllability.

cs.CV↗

arXiv2025

Evolving in Tasks: Empowering the Multi-modality Large Language Model as the Computer Use Agent

Computer use agents represent an emerging area in artificial intelligence, aiming to operate computers autonomously to fulfill user tasks, attracting significant attention from both industry and academia. However, the performance of existing agents remains insufficient for practical deployment. In this paper, we propose the Self-Evolution Agent (SEA) for computer operation, alongside three core innovations in data generation, reinforcement learning, and model enhancement to develop this agent. Specifically, we first design an automatic pipeline to generate verifiable task trajectories for training. Second, we propose Efficient Step-wise Reinforcement Learning to reduce the substantial computational overhead of long-horizon training. Finally, we introduce a model enhancement method that integrates grounding and planning capabilities into a single model without additional training. Leveraging these innovations, our SEA (with only 7B parameters) outperforms existing models of the same parameter scale and achieves performance comparable to larger models (e.g., 32B/72B parameters) on computer use tasks. We plan to release the model weights and related code as open-source resources in the future.

cs.AI↗

arXiv2025

Crafting Physical Adversarial Examples by Combining Differentiable and Physically Based Renders

Recently we have witnessed progress in hiding road vehicles against object detectors through adversarial camouflage in the digital world. The extension of this technique to the physical world is crucial for testing the robustness of autonomous driving systems. However, existing methods do not show good performances when applied to the physical world. This is partly due to insufficient photorealism in training examples, and lack of proper physical realization methods for camouflage. To generate a robust adversarial camouflage suitable for real vehicles, we propose a novel method called PAV-Camou. We propose to adjust the mapping from the coordinates in the 2D map to those of corresponding 3D model. This process is critical for mitigating texture distortion and ensuring the camouflage's effectiveness when applied in the real world. Then we combine two renderers with different characteristics to obtain adversarial examples that are photorealistic that closely mimic real-world lighting and texture properties. The method ensures that the generated textures remain effective under diverse environmental conditions. Our adversarial camouflage can be optimized and printed in the form of 2D patterns, allowing for direct application on real vehicles. Extensive experiments demonstrated that our proposed method achieved good performance in both the digital world and the physical world.

cs.GR↗

arXiv2024

Investigating triply heavy tetraquark states through QCD sum rules

We apply the method of QCD sum rules to study the $QQ\bar{Q}\bar{q}$ and $QQ\bar{Q}\bar{s}$ tetraquark states, where $Q=c,b$ and $q=u,d$, with the quantum number $J^P = 0^{+}$. We consider the contributions of vacuum condensates up to dimension-9 in the operator product expansion, and use the energy scale formula $\mu = \sqrt{M_{X}^2 - (i\mathbb{M}_c + j\mathbb{M}_b)^2} - k\mathbb{M}_s$ to determine the optimal energy scales for the QCD spectral densities. Our results indicate that triply charm tetraquark states $cc\bar{c}\bar{q}$ and $cc\bar{c}\bar{s}$ have masses in the ranges of $5.38-5.84\,\text{GeV}$ and $5.66-6.16\,\text{GeV}$, respectively. In the bottom sector, triply bottom tetraquark states $bb\bar{b}\bar{q}$ and $bb\bar{b}\bar{s}$ have masses in the ranges of $14.89-15.55\,\text{GeV}$ and $14.95-15.66\,\text{GeV}$, respectively. This study could help distinguish these states in upcoming high-energy nuclear and particle experiments.

hep-ph↗

arXiv2024

A novel configuration of gluonic tetraquark state

Inspired by the experimental measurement of the charmed hadronic state X(6900), we calculate the mass spectra of tetraquark hybrid states with configuration of $[8_{c}]_{Q\bar{Q}} \otimes [8_{c}]_{G} \otimes [8_{c}]_{Q\bar{Q}}$ in color, by virtue of QCD sum rules. The two feasible types of currents with quantum numbers $J^{PC} = 0^{++}$ and $0^{-+}$ are investigated, in which the contributions from operators up to dimension six are taken into account in operator product expansion (OPE). In the end, we find that, in charm sector, the tetracharm hybrid states with quantum number $0^{++}$ has a mass of about $6.98^{+0.16}_{-0.14} \, \text{GeV}$, while $0^{-+}$ state mass is about $7.26^{+0.16}_{-0.15} \, \text{GeV}$. The results overlap with the experimental observations, suggesting potential tetracharm hybrid interpretations. In bottom sector, calculation shows that the masses of tetrabottom hybrid states with quantum numbers $0^{++}$ and $0^{-+}$ are $19.30^{+0.16}_{-0.17} \, \text{GeV}$ and $19.50^{+0.17}_{-0.17} \, \text{GeV}$, respectively, which are left for future experimental confirmation.

hep-ph↗

arXiv2024

Enhancing binary classification: A new stacking method via leveraging computational geometry

Stacking, a potent ensemble learning method, leverages a meta-model to harness the strengths of multiple base models, thereby enhancing prediction accuracy. Traditional stacking techniques typically utilize established learning models, such as logistic regression, as the meta-model. This paper introduces a novel approach that integrates computational geometry techniques, specifically solving the maximum weighted rectangle problem, to develop a new meta-model for binary classification. Our method is evaluated on multiple open datasets, with statistical analysis showing its stability and demonstrating improvements in accuracy compared to current state-of-the-art stacking methods with out-of-fold predictions. This new stacking method also boasts two significant advantages: enhanced interpretability and the elimination of hyperparameter tuning for the meta-model, thus increasing its practicality. These merits make our method highly applicable not only in stacking ensemble learning but also in various real-world applications, such as hospital health evaluation scoring and bank credit scoring systems, offering a fresh evaluation perspective.

cs.LG↗

arXiv2024

Fully Charmed Tetraquark States in $8_{[c\bar{c}]}\otimes8_{[c\bar{c}]}$ Color Structure via QCD Sum Rules

Stimulated by the recent experimental results on the fully-charm tetraquark states, we systematically calculate the mass spectra of the fully-charm tetraquark states in $8_{[c\bar{c}]}\otimes8_{[c\bar{c}]}$ color configuration via QCD sum rules. By constructing nine $8_{[c\bar{c}]}\otimes8_{[c\bar{c}]}$ type currents with quantum numbers $J^{PC}=0^{-+},0^{--},1^{-+},1^{+-},1^{--}$ and $2^{++}$, we perform analytic calculation up to dimension six in the Operator Product Expansion (OPE). We find the fully-charm tetraquark states with $J^{PC}=1^{+-},2^{++}$ lie around 6.48 $\sim$ 6.62 GeV while the fully-charm tetraquark states with $J^{PC}=0^{-+},0^{--},1^{--},1^{-+}$ are about 6.85 $\sim$ 7.02 GeV. Notably, the mass predictions for the $c\bar{c}c\bar{c}$ tetraquarks, specifically those with $J^{PC}=2^{++}$, align with the broad structure identified by LHCb. Moreover, the masses of fully-charm tetraquarks with $J^{PC}=0^{-+}$ and $1^{-+}$ are anticipated to match closely with the mass of X(6900), considering the margin of error. Such findings hint at the presence of some $8_{[c\bar{c}]}\otimes8_{[c\bar{c}]}$ components within the di-$J/\psi$ structures observed by LHCb. The predictions for tetraquark states with $J^{PC}=0^{--},1^{+-},1^{--}$ may be accessible in the future BelleII, Super-B, PANDA, and LHCb experiments.

hep-ph↗

arXiv2024

Spectrum of $[8]_{[c\bar{s}]} \otimes [8]_{[q \bar{q^\prime}]}$ systems with quantum numbers $J^{P}=0^\pm$ and $1^\pm$

Inspired by the recent experimental progress on the $T_{c\bar{s}0}^a(2900)^{0/++}$, the fully open tetraquark spectrum with the configuration of $[8]_{[c\bar{s}]} \otimes [8]_{[q \bar{q^\prime}]}$ is systematically investigated by the QCD sum rules. In this article, we concentrate on the quantum numbers $J^{P}=0^{+}/0^{-}/1^{+}/1^{-}$. Firstly, we construct four scalar currents ($J^{P}=0^+$) in the form of $[8]_{[c\bar{s}]} \otimes [8]_{[u\bar{d}]}$ type tetraquark structure and perform the analysis of the QCD sum rules, where we consider the leading order contributions up to dimension 11 in the operator product expansion and retain contributions linear in the strange quark mass $m_s$. Our results, $M_{0^{+}}^A = 2.91_{-0.20}^{+0.20}~\text{GeV}$ and $M_{0^{+}}^C = 2.98_{-0.21}^{+0.20}~\text{GeV}$, are consistent with the experimentally discovered $T_{c\bar{s}0}^a(2900)^{++}$ within error margins. Thus, our calculation supports classifying $T_{c\bar{s}0}^a(2900)^{++}$ as a tetraquark state with the $[8]_{[c\bar{s}]} \otimes [8]_{u \bar{d}}$ color configuration. Moreover, on the basis of $J^P=0^+$, we also study tetraquark states with quantum numbers $J^P=0^-$, $1^+$, and $1^-$, predicting two new hadronic states awaiting experimental verification for each quantum number. Their masses are as follows: $M_{0^{-}}^{A}=3.45_{-0.10}^{+0.12}~\text{GeV}$, $M_{0^{-}}^{D}=3.20_{-0.19}^{+0.17}~\text{GeV}$, $M_{1^{+}}^{A}=2.95_{-0.19}^{+0.19}~\text{GeV}$, $M_{1^{+}}^{C}={2.95_{-0.19}^{+0.19}}~\text{GeV}$, $M_{1^{-}}^{A}={3.23_{-0.10}^{+0.11}}~\text{GeV}$ and $M_{1^{-}}^{D}={3.23_{-0.11}^{+0.09}}~\text{GeV}$. With advances in experimental techniques and accumulation of new data, these predicted results are hoped to be confirmed in future experiments.

hep-ph↗

arXiv2023

2023 Astrophotonics Roadmap: pathways to realizing multi-functional integrated astrophotonic instruments

Photonics offer numerous functionalities that can be used to realize astrophotonic instruments. The most spectacular example to date is the ESO Gravity instrument at the Very Large Telescope in Chile. Integrated astrophotonic devices stand to offer critical advantages for instrument development, including extreme miniaturization, as well as integration, superior thermal and mechanical stabilization owing to the small footprint, and high replicability offering cost savings. Numerous astrophotonic technologies have been developed to address shortcomings of conventional instruments to date, including for example the development of photonic lanterns, complex aperiodic fiber Bragg gratings, complex beam combiners to enable long baseline interferometry, and laser frequency combs for high precision spectral calibration of spectrometers. Despite these successes, the facility implementation of photonic solutions in astronomical instrumentation is currently limited because of (1) low throughputs from coupling to fibers, coupling fibers to chips, propagation and bend losses, device losses, etc, (2) difficulties with scaling to large channel count devices needed for large bandwidths and high resolutions, and (3) efficient integration of photonics with detectors, to name a few. In this roadmap, we identify 24 areas that need further development. We outline the challenges and advances needed across those areas covering design tools, simulation capabilities, fabrication processes, the need for entirely new components, integration and hybridization and the characterization of devices. To realize these advances the astrophotonics community will have to work cooperatively with industrial partners who have more advanced manufacturing capabilities. With the advances described herein, multi-functional instruments will be realized leading to novel observing capabilities for both ground and space platforms.

astro-ph.IM↗

arXiv2023

Angle-dependent pair production in the polarized two-photon Breit-Wheeler process

The advent of laser-driven high-intensity $\gamma$-photon beams has opened up new opportunities for designing advanced photon-photon colliders. Such colliders have the potential to produce a large yield of linear Breit-Wheeler (LBW) pairs in a single shot, which offers a unique platform for studying the polarized LBW process. In our recent work [Phys. Rev. D 105, L071902(2022)], we investigated the polarization characteristics of LBW pair production in CP $\gamma$-photon collisions. To fully clarify the polarization effects involving both CP and LP $\gamma$-photons, here we further investigate the LBW process using the polarized cross section with explicit azimuthal-angle dependence due to the base rotation of photon polarization vectors. We accomplished this by defining a new spin basis for positrons and electrons, which enables us to decouple the transverse and longitudinal spin components of $e^\pm$. By means of analytical calculations and Monte Carlo simulations, we find that the linear polarization of photon can induce the highly angle-dependent pair yield and polarization distributions. The comprehensive knowledge of the polarized LBW process will also open up avenues for investigating the higher-order photon-photon scattering, the laser-driven quantum electrodynamic plasmas and the high-energy astrophysics.

hep-ph↗

arXiv2023

Mass predictions of triply heavy hybrid baryons via QCD sum rules

In this article, we study the mass spectrum of the low-lying triply heavy hybrid baryon, which consists of three valence heavy quarks in a color octet and one valence gluon, with spin-parity $J^P=(\frac{1}{2})^+$ via QCD sum rules. This is the first study on the triply heavy hybrid baryons in the framework of QCD sum rules. After performing the QCD sum rule analysis, we find that the mass of $cccg$ hybrid baryon lies in $M_{cccg}= 5.91-6.13$ GeV. As a byproduct, the mass of the triply bottom hybrid baryon state is extracted to be around $M_{bbbg}=14.62-14.82$ GeV. The contributions up to dimension eight at the leading order of $\alpha_s$ (LO) in the operator product expansion are taken into account in the calculation. The triply charmed hybrid baryon predicted in this work can decay into one doubly charmed baryon and one charmed meson. Especially, we propose to search for $cccg$ hybrid baryon with $J^{P}= (1/2)^+$ in the P-wave decay channels $\Xi_{cc}^{++} D^0$, $\Xi_{cc}^{+} D^+$, and $\Xi_{ccs}^{+} D_s^+$, which may be accessible in future BelleII, Super-B, PANDA, and LHCb experiments.

hep-ph↗