arXiv subjects

Xiaoran Zhao

Publications and source records attributed to Xiaoran Zhao.

At least 19 recordsLinked to original sources

arXiv2026

Not Just What's There: Enabling CLIP to Comprehend Negated Visual Descriptions Without Fine-tuning

Vision-Language Models (VLMs) like CLIP struggle to understand negation, often embedding affirmatives and negatives similarly (e.g., matching "no dog" with dog images). Existing methods refine negation understanding via fine-tuning CLIP's text encoder, risking overfitting. In this work, we propose CLIPGlasses, a plug-and-play framework that enhances CLIP's ability to comprehend negated visual descriptions. CLIPGlasses adopts a dual-stage design: a Lens module disentangles negated semantics from text embeddings, and a Frame module predicts context-aware repulsion strength, which is integrated into a modified similarity computation to penalize alignment with negated semantics, thereby reducing false positive matches. Experiments show that CLIP equipped with CLIPGlasses achieves competitive in-domain performance and outperforms state-of-the-art methods in cross-domain generalization. Its superiority is especially evident under low-resource conditions, indicating stronger robustness across domains.

cs.CV↗

arXiv2024

LTOS: Layout-controllable Text-Object Synthesis via Adaptive Cross-attention Fusions

Controllable text-to-image generation synthesizes visual text and objects in images with certain conditions, which are frequently applied to emoji and poster generation. Visual text rendering and layout-to-image generation tasks have been popular in controllable text-to-image generation. However, each of these tasks typically focuses on single modality generation or rendering, leaving yet-to-be-bridged gaps between the approaches correspondingly designed for each of the tasks. In this paper, we combine text rendering and layout-to-image generation tasks into a single task: layout-controllable text-object synthesis (LTOS) task, aiming at synthesizing images with object and visual text based on predefined object layout and text contents. As compliant datasets are not readily available for our LTOS task, we construct a layout-aware text-object synthesis dataset, containing elaborate well-aligned labels of visual text and object information. Based on the dataset, we propose a layout-controllable text-object adaptive fusion (TOF) framework, which generates images with clear, legible visual text and plausible objects. We construct a visual-text rendering module to synthesize text and employ an object-layout control module to generate objects while integrating the two modules to harmoniously generate and integrate text content and objects in images. To better the image-text integration, we propose a self-adaptive cross-attention fusion module that helps the image generation to attend more to important text information. Within such a fusion module, we use a self-adaptive learnable factor to learn to flexibly control the influence of cross-attention outputs on image generation. Experimental results show that our method outperforms the state-of-the-art in LTOS, text rendering, and layout-to-image tasks, enabling harmonious visual text rendering and object generation.

cs.CV↗

arXiv2024

QCD corrections to the Golden decay channel of the Higgs boson

Future colliders aim to provide highly precise experimental measurements of the properties of the Higgs boson. In order to benefit from these precision machines, theoretical errors in the Higgs sector observables have to match at least the experimental uncertainties. The theoretical uncertainties in the Higgs sector observables can be reduced by including missing higher-order terms in their perturbative calculations. In this direction, we compute the mixed QCD-electroweak corrections at ${\mathcal O}(αα_s)$ to the Higgs decay into four charged leptons by considering the golden decay channel, $ H \to e^+e^-μ^+μ^-$. Due to color conservation, these corrections receive contributions only from the two-loop virtual diagrams. In the complex mass scheme, we find that the mixed QCD-electroweak corrections to the partial decay width, relative to the leading order predictions, are positive and about $0.27\%$ for $α_s(M_Z)$. Relative to the next-to-leading order electroweak corrections, the mixed QCD-electroweak corrections are found to be approximately $18\%$ for $α_s(M_Z)$. With respect to the leading order, we observe a flat effect of the mixed QCD-electroweak corrections on the invariant mass distribution of the lepton pairs with fixed QCD coupling. The $ϕ$ distribution, due to the mixed QCD-electroweak corrections, follows a $(1-\cos ϕ)$ dependence.

hep-ph↗

arXiv2023

Constraining rare B decays by $μ^+μ^-\to tc$ at future lepton colliders

Motivated by the recent rare B decays measurements, we study the matching procedure of operators $O_9, O_{10}$ in the low energy effective Hamiltonian and operators in the Standard Model effective theory (SMEFT). It is noticed that there are more related operators in the SMEFT whose coefficients can not be determined only from the low-energy data from B physics. We demonstrate how to determine these coefficients with some new physics models, like $Z^\prime$ model and leptoquark models, and then consider how to probe these operators of SMEFT at high energy by using the process $μ^+μ^-\to tc$ at future muon colliders, which can provide complementary information except for $μ^+ μ^- \to b s$ on the underlying models which lead to rare B decay processes. We perform a Monte Carlo study (a hadron level analysis) to show how to separate the signal events from the SM background events and estimate the sensitivity to the Wilson coefficients for different models.

hep-ph↗

arXiv2022

Going all the way in the search for WIMP dark matter at the muon collider through precision measurements

Dark Matter is a necessary ingredient for a complete theory of Nature, which has so far remained elusive in laboratory searches for new particles. Searches at current and future colliders are in principle a promising way to search for electroweak charged dark matter particles, but the sensitivity of experiments at the LHC and future $pp$ colliders falls short to fully probe the whole mass range allowed for electroweak charged dark matter particles, which extends in principle up to the O(0.1) PeV. In this work we examine the effect of on-shell and off-shell propagation of electroweak charged thermal dark matter particles on integrated and differential rates of several Standard Model final states at the muon collider, considering candidates from weak 2-plet at the TeV scale up to 7-plet and 9-plet in the O(0.1) PeV ballpark. For fermionic WIMPs we find that all dark matter candidates with $n\leq 5$, corresponding to a thermal mass up to 14~TeV, can be excluded at the high-energy muon collider for some center-of-mass energy at or below 14~TeV. For the $n>5$ WIMPs our results show that higher energy muon colliders offer a route to conclusively probe both scalar and fermionic WIMPs off-shell production all the way up to the perturbativity bound for WIMP dark matter at O(0.1)~PeV. Our results bring WIMPs over the whole allowed mass range in the realm of collider searches and motivate research and development for the realization of a high energy muon collider.

hep-ph↗

arXiv2022

Shifts in BCFW method for QED

We study the application of BCFW recursion relations to the QED processes $0\to e^- e^+ n γ$. Based on 6-point amplitudes (both MHVA and NMHVA) computed from Feynman diagrams in the Berends-Giele gauge, we conduct a comprehensive study on all different shifts. Then we propose a new shift (LLYZ shift) which can lead to the full amplitudes for these processes and can have some realistic computation advantages. We compare the number of terms and the independent amplitudes of this novel shift with a few typical shifts.

hep-th↗

arXiv2022

On the NLO QCD Corrections to Gluon-Initiated ZH Production

We compute the QCD corrections at next-to-leading order for the process $gg \rightarrow ZH$, including both the virtual two-loop terms and real-emission contributions. The two-loop box diagrams in the virtual corrections are approximated analytically over the complete phase space, combining the results of an expansion in the limit of small transverse momentum and an expansion in the regime of high energy. We obtain both inclusive and differential results for the cross section. We find that the NLO QCD corrections are of the same size as the LO contribution up to $ZH$ invariant masses close to 1 TeV, but they increase significantly when higher energies are considered, due to a class of real-emission diagrams in which the $Z$ boson is radiated from an open quark line. Finally, we estimate the uncertainty due to the renormalization scheme used for the top-quark mass both on the total and differential cross section.

hep-ph↗

arXiv2021

Lepton collisions in MadGraph5_aMC@NLO

MadGraph5_aMC@NLO is a software package that allows one to simulate processes of arbitrary complexity, at both the leading and the next-to-leading order perturbative accuracy, with or without matching and multi-jet merging to parton showers. It has been designed for, and so far primarily employed in the context of, hadronic collisions. In this note, we document the implementation of a few technical features that are necessary to extend its scope to realistic ee collider environments. We limit ourselves to discussing the unpolarized beam case, but we point out that the treatment of polarized beams is conceptually identical, and that the structure we set up can easily be extended to carry out simulations at muon colliders.

hep-ph↗

arXiv2021

Highly Boosted Higgs Bosons and Unitarity in Vector-Boson Fusion at Future Hadron Colliders

We study the observability of new interactions which modify Higgs-pair production via vector-boson fusion processes at the LHC and at future proton-proton colliders. In an effective-Lagrangian approach, we explore in particular the effect of the operator $h^2 W_{μν}^a W^{a,μν}$, which describes the interaction of the Higgs boson with transverse vector-boson polarization modes. By tagging highly boosted Higgs bosons in the final state, we determine projected bounds for the coefficient of this operator at the LHC and at a future 27 TeV or 100 TeV collider. Taking into account unitarity constraints, we estimate the new-physics discovery potential of Higgs pair production in this channel.

hep-ph↗

arXiv2020

Measuring the quartic Higgs self-coupling at a multi-TeV muon collider

Measuring the shape of the Higgs boson potential is of paramount importance and will be a challenging task at current as well as future colliders. While the expectations for the measurement of the trilinear Higgs self-coupling are rather promising, an accurate measurement of the quartic self-coupling interaction is presently considered extremely challenging even at a future 100 TeV proton-proton collider. In this work we explore the sensitivity that a muon collider with a center of mass energy in the multi-TeV range, and luminosities of the order of 10^35cm^-2s^-1, as presently under discussion, might provide thanks to a rather large three Higgs-boson production and to a limited background. By performing a first and simple analysis, we find a clear indication that a muon collider could provide a determination of the quartic Higgs self-coupling that is significantly better than what is currently considered attainable at other future colliders.

hep-ph↗

arXiv2020

Vector boson fusion at multi-TeV muon colliders

High-energy lepton colliders with a centre-of-mass energy in the multi-TeV range are currently considered among the most challenging and far-reaching future accelerator projects. Studies performed so far have mostly focused on the reach for new phenomena in lepton-antilepton annihilation channels. In this work we observe that starting from collider energies of a few TeV, electroweak (EW) vector boson fusion/scattering (VBF) at lepton colliders becomes the dominant production mode for all Standard Model processes relevant to studying the EW sector. In many cases we find that this also holds for new physics. We quantify the size and the growth of VBF cross sections with collider energy for a number of SM and new physics processes. By considering luminosity scenarios achievable at a muon collider, we conclude that such a machine would effectively be a "high-luminosity weak boson collider," and subsequently offer a wide range of opportunities to precisely measure EW and Higgs coupling as well as to discover new particles.

hep-ph↗

arXiv2019

Multi-Higgs Production and Unitarity in Vector-Boson Fusion at Future Hadron Colliders

We study multi-Higgs final states in vector boson fusion (VBF) processes at the LHC and at future proton-proton colliders, focusing on the prospects for measurements at 27 TeV and at 100 TeV. We use an effective Lagrangian which includes higher-dimensional operators in the mass eigenstates which are relevant to VBF processes and relate this to specific parameterizations and models for new physics in the Higgs sector. We derive theoretical constraints on the parameter space from the unitarity of $2\to n$ scattering amplitudes and apply the results to $V V \to hh$ and $h h h$ processes, where $V=W,Z$. As a result, we present constraints on differential distributions as appropriate to the study of $V V \to hh$ and $h h h$ processes.

hep-ph↗

arXiv2019

Evaluating multi-loop Feynman integrals numerically through differential equations

The computation of Feynman integrals is often the bottleneck of multi-loop calculations. We propose and implement a new method to efficiently evaluate such integrals in the physical region through the numerical integration of a suitable set of differential equations, where the initial conditions are provided in the unphysical region via the sector decomposition method. We present numerical results for a set of two-loop integrals, where the non-planar ones complete the master integrals for $gg\toγγ$ and $q\bar{q}\toγγ$ scattering mediated by the top quark.

hep-ph↗

arXiv2019

Bootstrapping solutions of scattering equations

The scattering equations are a set of algebraic equations connecting the kinematic space of massless particles and the moduli space of Riemann spheres with marked points. We present an efficient method for solving the scattering equations based on the numerical algebraic geometry. The cornerstone of our method is the concept of the physical homotopy between different points in the kinematic space, which naturally induces a homotopy of the scattering equations. As a result, the solutions of the scattering equations with different points in the kinematic space can be tracked from each other. Finally, with the help of soft limits, all solutions can be bootstrapped from the known solution for the four-particle scattering.

hep-th↗

arXiv2018

Top-quark effects in diphoton production through gluon fusion at NLO in QCD

At hadron colliders, the leading production mechanism for a pair of photons is from quark-anti-quark annihilation at the tree level. However, due to large gluon-gluon luminosity, the loop-induced process $gg\to γγ$ provides a substantial contribution. In particular, the amplitudes mediated by the top quark become important at the $t \bar t$ threshold and above. In this letter we present the first complete computation of the next-to-leading order (NLO) corrections (up to $α_S^3$) to this process, including contributions from the top quark. These entail two-loop diagrams with massive propagators whose analytic expressions are unknown and have been evaluated numerically. We find that the NLO corrections to the top-quark induced terms are very large at low diphoton invariant mass $m(γγ)$ and close to the $t \bar t$ threshold. The full result including five massless quarks and top quark contributions at NLO displays a much more pronounced change of slope in the $m(γγ)$ distribution at $t \bar t$ threshold than at LO and an enhancement at high invariant mass with respect to the massless calculation.

hep-ph↗

arXiv2018

Probing the scalar potential via double Higgs boson production at hadron colliders

We present a sensitivity study on the cubic and quartic self couplings in double Higgs production via gluon fusion at hadron colliders. Considering the relevant operators in the Standard Model Effective Field Theory up to dimension eight, we calculate the dominant contributions up to two-loop level, where the first dependence on the quartic interaction appears. Our approach allows to study the independent variations of the two self couplings and to clearly identify the terms necessary to satisfy gauge invariance and to obtain UV-finite results order by order in perturbation theory. We focus on the $b \bar b γγ$ signature for simplicity and provide the expected bounds for the cubic and quartic self couplings at the 14 TeV LHC with 3000 fb$^{-1}$ (HL-LHC) and for a future 100 TeV collider (FCC-100) with 30 ab$^{-1}$. We find that while the HL-LHC will provide very limited sensitivity on the quartic self coupling, precision measurements of double Higgs production at a FCC-100 will offer the opportunity to set competitive bounds. We show that combining information from double and triple Higgs production leads to significantly improved prospects for the determination of the quartic self coupling.

hep-ph↗

arXiv2018

Probing the Higgs boson-gluon coupling via the jet energy profile at $e^+e^-$ colliders

The effective coupling of the Higgs boson to a gluon pair is one of the most important parameters to test the Standard Model and search for the new physics beyond. In this paper, we propose several new observables based on the jet energy profile to extract the effective coupling. The statistical uncertainties of the effective coupling extracted by using new observables are derived and estimated based on the simulation at the future $e^+e^-$ collider for $250$ GeV center-of-mass energy and 5 ab$^{-1}$ integrated luminosity. We found that the statistical uncertainties of effective coupling via the optimized observable can reach about $1.6\%$ in the channels of a $Z$ boson decaying to lepton pairs and is reduced by $52\%$ compared to the relevant uncertainties in the conventional approach. These new observables potentially can be helpful for the measurement of effective coupling at future $e^+e^-$ colliders.

hep-ph↗

arXiv2018

Constraining the Higgs self couplings at $e^+e^-$ colliders

We study the sensitivity to the shape of the Higgs potential of single, double, and triple Higgs production at future $e^+e^-$ colliders. Physics beyond the Standard Model is parameterised through the inclusion of higher-dimensional operators $(Φ^\dagger Φ- v^2/2)^n/Λ^{(2n-4)}$ with $n=3,4$, which allows a consistent treatment of independent deviations of the cubic and quartic self couplings beyond the tree level. We calculate the effects induced by a modified potential up to one loop in single and double Higgs production and at the tree level in triple Higgs production, for both $Z$ boson associated and $W$ boson fusion production mechanisms. We consider two different scenarios. First, the dimension six operator provides the dominant contribution (as expected, for instance, in a linear effective-field-theory(EFT)); we find in this case that the corresponding Wilson coefficient can be determined at $\mathcal{O}(10\%)$ accuracy by just combining accurate measurements of single Higgs cross sections at $\sqrt{\hat s}=$240-250 GeV and double Higgs production in $W$ boson fusion at higher energies. Second, both operators of dimension six and eight can give effects of similar order, i.e., independent quartic self coupling deviations are present. Constraints on Wilson coefficients can be best tested by combining measurements from single, double and triple Higgs production. Given that the sensitivity of single Higgs production to the dimension eight operator is presently unknown, we consider double and triple Higgs production and show that combining their information colliders at higher energies will provide first coarse constraints on the corresponding Wilson coefficient.

hep-ph↗