Searcharxiv⌕ Search

arXiv subjects

Ze Zhou

Publications and source records attributed to Ze Zhou.

At least 19 recordsLinked to original sources

LeafSkill: Efficient Leaf-Level Action Grounding for Recursive LLM Agents

Recursive large language model (LLM) agents solve long-horizon tasks through decomposition, execution, and environmental feedback. Yet after producing an atomic subtask, they often invoke the same large model again to translate a familiar instruction into an executable action. We observe that high-level planning is task-specific, whereas leaf-level action grounding recurs across tasks and can be checked against environment-admitted actions. We introduce LeafSkill, an external reuse layer that retrieves semantically related grounding experience, binds action templates to current objects and admissible actions, and uses a lightweight 1.7B selector to choose a bound action or defer the unchanged subtask to the original agent. Skills remain outside the model parameters, allowing a frozen selector to use post-training additions. Under lexical and syntactic paraphrases, exact matching retrieves none of the reference actions, while semantic retrieval preserves 96.0% and 97.0% candidate coverage; the frozen selector adopts 97.34% of unseen skill types. On an independent confirmation set of 200 Robotouille tasks, LeafSkill completes 137 tasks versus 121 for ReCAP (68.5% vs. 60.5%, exact McNemar p=.0259). On the 106 tasks completed by both, it reduces Qwen3.5-27B tokens by 14.42% and calls by 14.08%. A separate held-out repeat yields a consistent 9.0-point success gain and 18.98% token reduction. Reusable leaf grounding can thus improve task completion while reducing host-model inference without replacing high-level planning.

cs.AI↗

Morse theory and moduli spaces of simple polylines and polygons

We prove that a smooth $d$-manifold $M$ is diffeomorphic to $\mathbb R^d$ if it admits a Lyapunov-Reeb function, i.e., a smooth map $f\colon M\to\mathbb R$ that is proper, lower-bounded, and possesses exactly one critical point. We further establish that the moduli spaces of simple polylines and positively oriented polygons with fixed or varying side lengths are diffeomorphic to Euclidean spaces, by explicitly constructing Lyapunov-Reeb functions on these spaces. This yields a refined resolution to the Carpenter's Rule Problem and settles a conjecture proposed by González and Sedano-Mendoza. In addition, we describe foliation structures of the moduli spaces and develop algorithms for the refolding problem.

math.CO↗

The form factors of transition process $\mathcal{B}_{Q_{1}Q_{2}}(\frac{1}{2}^{+})\rightarrow\mathcal{B}_{Q_{1}}^{*}(\frac{3}{2}^{+})$ and their applications to semileptonic decays

In the framework of QCD sum rules, we obtain the form factors of the weak transition process $\mathcal{B}_{Q_{1}Q_{2}}(\frac{1}{2}^{+})\rightarrow\mathcal{B}_{Q_{1}}^{*}(\frac{3}{2}^{+})$. In this analysis, we eliminate the interferences of the low spin states and negative parity states to final results by employing 16 different dirac structures. As an application, these form factors are finally used to analyze the semileptonic decays of $\mathcal{B}_{Q_{1}Q_{2}}(\frac{1}{2}^{+})\rightarrow\mathcal{B}_{Q_{1}}^{*}(\frac{3}{2}^{+})lν$, where these decays are driven by the transition processes $c\rightarrow d/s+l^{+}+ν_{l}$ and $b\rightarrow u+l^{-}+\overlineν_{l}$. The predicted physical quantities include not only the partial widths, ratios of $Γ_{L}/Γ_{T}$ and the branching fractions, but also some observables such as the forward-backward asymmetry parameter $A_{FB}^{l}$ of lepton, the $P_z^{F}$ component of the polarization vector for daughter baryon and the longitudinal polarization of the lepton $P_z^{l}$. We hope all of these theoretical predictions about the weak decays will be helpful for studying the properties of doubly heavy baryons in experiments in the future.

hep-ph↗

GREAT-ESDD: Graph Edge Attention Network for Effectively Steered Delivery Decisions

Online, same-day delivery requires operators to constantly update their routes, and the routing cost for deliveries increases with the number of customers. Out-of-home (OOH) delivery options can help companies handle increasing customer numbers while keeping delivery costs low. We develop a dynamic pricing-and-routing framework for this online ordering scenario with home and OOH delivery options. First, we develop a neural solver for the capacitated vehicle routing problem (CVRP) based on the Graph-Edge-Attention-Network (GREAT). We train it to solve CVRP with varying instance sizes and starting conditions, allowing it to reoptimize partially executed routes under changing conditions. It is fast while achieving solution quality competitive with traditional baselines. Our solver processes several CVRP instances in parallel in batches and we use it to solve the candidate routing problems that arise from an incoming customer's choice of home or OOH delivery. This provides an estimate of the delivery cost increase per option, which we use to set dynamic prices. Further, it directly provides the updated route once the customer decides. We trade off delivery fees against routing costs and steer customers toward OOH delivery when home delivery would be costly. However, effective steering requires us to know the customer preferences. To learn those online, we use Thompson sampling with a Laplace-approximated posterior. We call the resulting framework GREAT-ESDD (GREAT for Effectively Steered Delivery Decisions). Compared to a traditional routing solver, it increases profit by around 11% under a similar pricing decision time budget. Moreover, we show that we can estimate customer preferences successfully, outperforming a static pricing strategy after only 5 days of simulated operations.

cs.AI↗

Assortment Control Unlocks the Value of Dynamic Pricing in Mixed Last-Mile Delivery

The growth of e-commerce has intensified the need for last-mile delivery systems that can jointly manage customer choice and operational efficiency. We study the Dynamic Offering and Pricing of Mixed Delivery Options problem, in which a logistics service provider dynamically selects and prices attended home-delivery time slots and out-of-home pickup options for sequentially arriving customers. Each decision affects immediate revenue, customer acceptance, and the route-dependent fulfillment cost realized at the end of the booking horizon. We formulate DOPMDO as a finite-horizon Markov decision process and propose State-Value Anchored Pricing via Approximate Dynamic Programming. The method learns a continuation-value approximation on an aggregate state representation of the mixed-delivery system and uses accepted-versus-rejected value differences to estimate option-level opportunity costs. These opportunity costs are embedded in an anchored trust-region pricing problem around a calibrated fine-static benchmark. Computational experiments on the real-world Seattle instance show that SVAP--ADP increases mean episode profit by 7.0\% relative to current practice (95\% CI: 6.7--7.2\%), primarily by reducing terminal fulfillment cost while maintaining a stable home--locker--opt-out mix. Assortment-control experiments show that dynamic pricing is most effective when the menu exposes operationally valuable locker alternatives, with richer candidate pools delivering substantially larger gains than restricted nearest-locker menus. These results indicate that anticipatory pricing and assortment control are complementary: pricing steers customers toward lower-cost options, but the menu determines whether high-value consolidation opportunities are available in the first place.

math.OC↗

A multi-stage soft computing framework for complex disease modelling and decision support: A liver cirrhosis case study

Liver cirrhosis is a major global health problem causing millions of deaths annually, and timely detection with aggressive treatment can significantly improve patients' quality of life. Modelling complex diseases from biomedical data is computationally challenging due to high dimensionality, strong feature correlations, noise, and limited labelled samples. Conventional Machine Learning (ML) pipelines often struggle with robustness, interpretability, and generalisation under such conditions. In this study, we propose an ML-driven multi-stage decision framework for complex disease modelling and therapeutic exploration. The framework integrates single-cell transcriptomic profiling, high-dimensional network-based feature stabilisation, multi-model learning, deep representation construction, and post-hoc decision support. Specifically, single-cell sequencing data were analysed to identify key cellular subpopulations, followed by high-dimensional weighted gene co-expression network analysis (hdWGCNA) to stabilise gene modules under sparsity and noise. To enhance non-linear feature interaction modelling, tabular molecular features were restructured into two-dimensional disease maps and analysed using a CNN. Finally, molecular docking was incorporated as a decision-support module to evaluate candidate therapeutic compounds. Using liver cirrhosis as a representative case, the framework identified a disease-associated endothelial subpopulation and extracted seven robust signature genes (HSPB1, GADD45A, CLDN5, ATP1B3, C1QBP, ENPP2, and PARL). The CNN-based representation learning module outperformed conventional pipelines in classification. The framework is disease-agnostic and readily extends to other omics-driven biomedical applications involving uncertainty, heterogeneity, and limited samples.

q-bio.OT↗

Analysis of the semileptonic decays of $Ξ_{cc}$ and $Ω_{cc}$ baryons in QCD sum rules

We firstly carry out a systematic analysis on the spin $\frac{1}{2}^{+}\rightarrow\frac{3}{2}^{+}$ weak transition process in the framework of three-point QCD sum rules, where the initial and final states are doubly and singly charmed baryons. In the phenomenological side, all possible couplings of interpolating current to hadronic states are considered. In doing operator production expansion at QCD side, the contributions of the perturbative part, vacuum condensate terms of $\langle{\bar qq}\rangle$, $\langle g_{s}^{2}GG\rangle$, $\langle \bar q g_{s}σGq\rangle$ and $g_{s}^{2}\langle{\bar qq}\rangle^{2}$ are all considered. After the form factors in space-like region ($Q^2>0$) are obtained, the numerical results are extrapolated into time-like region ($Q^2<0$) by a fitting function. Using the predicted form factors, we finally analyze the semileptonic decays of $Ξ_{cc}^{++}\rightarrow Σ_{c}^{*+}l^{+}ν_{l}$, $Ξ_{cc}^{++}\rightarrow Ξ_{c}^{\prime*+}l^{+}ν_{l}$, $Ω_{cc}^{+}\rightarrowΞ_{c}^{\prime*0}l^{+}ν_{l}$ and $Ω_{cc}^{+}\rightarrow Ω_{c}^{*0}l^{+}ν_{l}$ with $l=e,μ$. The predictions in this work can deepen our understanding of the dynamics in the decay processes of doubly heavy baryons and provide useful information to explore the possibility of new physics in heavy baryonic decay channels.

hep-ph↗

A gradient descent algorithm for computing circle patterns

This paper presents a new algorithm for generating planar circle patterns. The algorithm employs gradient descent and conjugate gradient method to compute circle radii and centers separately. Compared with existing algorithms, the proposed method is more efficient in computing centers of circles and is applicable for realizing circle patterns with possible obtuse overlap angles.

cs.CG↗

Analysis of the strong vertices of hadronic molecules $DK$, $D^*K$, $DK^*$ and their bottom analogs

In this work, we analyze the strong vertices of hadronic molecules $DK$, $D^*K$, $DK^*$ and their bottom analogs within the framework of three-point QCD sum rules. The coupling between interpolating currents and low spin particles is considered in the phenomenological side, and the vacuum condensates $\left\langle \bar qq \right\rangle ,\left\langle g_s^2GG \right\rangle ,\left\langle \bar qg_sσGq \right\rangle ,\left\langle g_s^3GGG \right\rangle ,{\left\langle \bar qq \right\rangle ^2}$ are included in the QCD side. As an application of strong coupling constants, we also obtain the partial decay widths of these states, where $Γ_{T_{DK}\rightarrow D_sπ}=10.3_{-2.9}^{+3.0}\mathrm{KeV}$, $Γ_{T_{D^*K}\rightarrow D_s^*π}=24.8_{-5.4}^{+5.5}\mathrm{KeV}$, $Γ_{T_{BK}\rightarrow BK}=101_{-21}^{+36}\mathrm{MeV}$ and $Γ_{T_{B^*K}\rightarrow B^*K}=142_{-25}^{+52}\mathrm{MeV}$. It is shown that the results of $Γ_{T_{DK}\rightarrow D_sπ}$ and $Γ_{T_{D^*K}\rightarrow D_s^*π}$ are compatible with the experimental data of $D_{s0}^*(2317)$ and $D_{s1}(2460)$.

hep-ph↗

Analysis of the strong vertices $Λ_cD^{(*)}N^*(1535)$ and $Λ_bB^{(*)}N^*(1535)$ in QCD sum rules

In this article, we firstly analyze the mass and pole residue of negative parity nucleon $N^*(1535)$ within the two-point QCD sum rules. Basing on these results, we continuously study the strong coupling constants of vertices $Λ_cDN^*$, $Λ_cD^*N^*$, $Λ_bBN^*$ and $Λ_bB^*N^*$ in the framework of three-point QCD sum rules. At hadron side, all possible couplings of interpolating current to hadronic states are considered. At QCD side, the contributions of vacuum condensate terms $\langle\bar{q}q\rangle$, $\langle g_s^2GG\rangle$, $\langle\bar{q} g_sσGq\rangle$, $\langle\bar{q}q\rangle^2$ and $g_s^2\langle\bar{q}q\rangle^2$ are also considered. By setting the four momentum of $D^{(*)}[B^{(*)}]$ mesons off-shell, the strong coupling constants in deep space-like regions ($Q^2=-q^2\ggΛ_{QCD}^2$) are obtained. Then, the momentum dependent coupling constants in space-like regions are fitted into analytical function $G(Q^2)$ and are extrapolated into time-like regions ($Q^2<0$). Finally, the on-shell values of strong coupling constants are obtained by taking $Q^{2}=-m_{D^{(*)}[B^{(*)}]}^2$. The results are $G_{Λ_cDN^*}(Q^2=-m_D^2)=4.06^{+0.96}_{-0.75}$, $f_{Λ_cD^*N^*}(Q^2=-m_{D^*}^2)=3.73^{+0.68}_{-0.16}$, $g_{Λ_cD^*N^*}(Q^2=-m_{D^*}^2)=9.22^{+3.16}_{-0.36}$, $G_{Λ_bBN^*}(Q^2=-m_B^2)=9.11^{+1.54}_{-1.61}$, $f_{Λ_bB^*N^*}(Q^2=-m_{B^*}^2)=8.55^{+2.69}_{-2.21}$ and $g_{Λ_bB^*N^*}(Q^2=-m_{B^*}^2)=-0.25^{+0.16}_{-0.01}$.

hep-ph↗

Analysis of the hadronic molecules $DK$, $D^*K$, $DK^*$ and their bottom analogs with QCD sum rules

In this work, we construct the color-singlet-color-singlet type currents to study the masses and pole residues of charm-strange tetraquark states and their bottom analogs with $J^P$ = $0^+$ and $1^+$ by using two-point QCD sum rules, where the vacuum condensates are considered up to dimension 12. The predicted masses for $DK$, $D^*K$ and $DK^*$ molecular states are $2.322_{ - 0.072}^{ + 0.066}$ GeV, $2.457_{ - 0.068}^{ + 0.064}$ GeV and $2.538_{ - 0.062}^{ + 0.059}$ GeV. These results are consistent well with the experimental data of $D_{s0}(2317)$, $D_{s1}(2460)$ and ${D}_{s1}(2536)$, respectively. The theoretical results for $BK$ and $B^*K$ molecular states are $5.970_{ - 0.064}^{ + 0.061}$ GeV and $6.050_{ - 0.064}^{ + 0.062}$ GeV which are all higher than their own thresholds. Finally, the mass of hadronic molecule $BK^*$ is predicted to be $6.158_{ - 0.063}^{ + 0.061}$ GeV. This value is lower than the threshold of $BK^*$, which implies that it may be a bound hadronic molecular state.

hep-ph↗

Systematic analysis of the mass spectra of triply heavy baryons

The mass spectra, root mean square (r.m.s.) radii and radial density distributions of $Ω_{ccb}$ and $Ω_{bbc}$ baryons are firstly analyzed in the present work. The calculations are carried out in the frame work of relativized quark model, where the baryon is regarded as a real three-quark system. Our results show that the excited energy of charmed-bottom triply baryons are always associated with heavier quark. This means the lowest state of $Ω_{ccb}$ baryon is dominated by the $λ$-mode, however, the dominant orbital excitation for $Ω_{bbc}$ baryon is $ρ$-mode. In addition, the influence of configuration mixing on mass spectrum, which is induced by different angular momentum assignments, is also analyzed. It shows that energy of the lowest state will be further lowered by this mixing effect. According to this conclusion, we systematically analyze the mass spectra of the ground and excited states($1S\sim4S$, $1P\sim4P$, $1D\sim4D$, $1F\sim4F$ and $1G\sim4G$) of $Ω_{ccb}$, $Ω_{bbc}$, $Ω_{ccc}$ and $Ω_{bbb}$ baryons. Finally, with the predicated mass spectra, the Regge trajectories of these heavy baryons in the ($J$,$M^{2}$) plane are constructed.

hep-ph↗

CL-MoE: Enhancing Multimodal Large Language Model with Dual Momentum Mixture-of-Experts for Continual Visual Question Answering

Multimodal large language models (MLLMs) have garnered widespread attention from researchers due to their remarkable understanding and generation capabilities in visual language tasks (e.g., visual question answering). However, the rapid pace of knowledge updates in the real world makes offline training of MLLMs costly, and when faced with non-stationary data streams, MLLMs suffer from catastrophic forgetting during learning. In this paper, we propose an MLLMs-based dual momentum Mixture-of-Experts (CL-MoE) framework for continual visual question answering (VQA). We integrate MLLMs with continual learning to utilize the rich commonsense knowledge in LLMs. We introduce a Dual-Router MoE (RMoE) strategy to select the global and local experts using task-level and instance-level routers, to robustly assign weights to the experts most appropriate for the task. Then, we design a dynamic Momentum MoE (MMoE) to update the parameters of experts dynamically based on the relationships between the experts and tasks/instances, so that the model can absorb new knowledge while maintaining existing knowledge. The extensive experimental results indicate that our method achieves state-of-the-art performance on 10 VQA tasks, proving the effectiveness of our approach.

cs.CV↗

Analysis of the form factors of $B_c\rightarrow D^{(*)}$, $D_{s}^{(*)}$ and their nonleptonic decays

This article is devoted to calculating the form factors of $B_c \to D^{*}$, $B_c \to D$, $B_c \to D_s^{*}$ and $B_c \to D_s$ transitions in the framework of three-point QCD sum rules. At the QCD side, the contributions of $\langle\overline{q}q\rangle$, $\langle\overline{q}g_{s}σGq\rangle$, $\langle g_{s}^{2}G^{2}\rangle$, $\langle f^{3}G^{3}\rangle$ and $\langle\overline{q}q\rangle \langle g_{s}^{2}G^{2}\rangle$ are taken into account. With the obtained form factors, the decay widths and branching ratios of several two-body nonleptonic decay processes $B_c \to η_c D^{*}$, $η_c D$, $ J/ψD^{*}$, $ J/ψD$, $η_c D_s^{*}$, $η_c D_s$, $J/ψD_s^{*}$ and $J/ψD_s$ are predicted. These results about the form factors and decay properties of $B_c$ meson provide useful information for us to study the heavy-quark dynamics.

hep-ph↗

The ground states of hidden-charm tetraquarks and their radial excitations

Inspired by the great progress in the observations of charmonium-like states in recent years, we perform a systematic analysis about the ground states and the first radially excited states of $qc\bar{q}\bar{c}$ ($q$=$u/d$ and $s$) tetraquark systems. Their mass spectra, root mean square (r.m.s.) radii and radial density distributions are predicted within the framework of relativized quark model. By comparing with experimental data, some potential candidates for hidden-charm tetraquark states are suggested. For $qc\bar{q}\bar{c}$ ($q$=$u/d$) system, if $Z_{c}(3900)$ is supposed to be a compact tetraquark state with $J^{PC}=1^{+-}$, $Z(4430)$ can be interpreted as the first radially excited states of $Z_{c}(3900)$. Another broad structure $Z_{c}(4200)$ can also be explained as a partner of $Z_{c}(3900)$, and it arise from a higher state with $J^{PC}=1^{+-}$. In addition, theoretical predictions indicate that the possible assignments for $X(3930)$, $X(4050)$ and $X(4250)$ are low lying $0^{++}$ tetraquark states. As for the $sc\bar{s}\bar{c}$ system, $X(4140)$ and $X(4274)$ structures can be interpreted as this type of tetraquark states with $J^{PC}=1^{++}$, and $X(4350)$ can be described as a $sc\bar{s}\bar{c}$ tetraquark with $J^{PC}=0^{++}$. With regard to $qc\bar{s}\bar{c}$ ($q$=$u/d$) system, we find two potential candidates for this type of tetraquark, which are $Z_{cs}(4000)$ and $Z_{cs}(4220)$ structures. The measured masses of these two structures are in agreement with theoretical predictions for the $1^{+}$ state.

hep-ph↗

Generalizing Andreev's Theorem via circle patterns

In this paper we derive an extended Circle Pattern Theorem that allows obtuse overlap angles. As a consequence, we characterize a subclass of compact convex hyperbolic polyhedra with possibly obtuse dihedral angles and thus generalize Andreev's Theorem. For proofs of these results, we establish a discrete analog of the "weak solution/regularity theory" scheme.

math.GT↗

Polygons inscribed in Jordan curves with prescribed edge ratios

Let $J$ be a simple closed curve in $\mathbb R^{k}$ $(k\geq2)$ that is differentiable with non-zero derivative at a point $A_0\in J$. For a tuple of positive reals $a_1,\cdots,a_n$ $(n\geq3)$, each of which is less than the sum of the others, we show that there exists a polygon $Q_n$ inscribed in $J$ with sides of lengths proportional to $(a_1,\cdots,a_n)$. As a consequence, we prove the existence of triangle inscribed in $J$ similar to any given triangle.

math.GT↗

EduChat: A Large-Scale Language Model-based Chatbot System for Intelligent Education

EduChat (https://www.educhat.top/) is a large-scale language model (LLM)-based chatbot system in the education domain. Its goal is to support personalized, fair, and compassionate intelligent education, serving teachers, students, and parents. Guided by theories from psychology and education, it further strengthens educational functions such as open question answering, essay assessment, Socratic teaching, and emotional support based on the existing basic LLMs. Particularly, we learn domain-specific knowledge by pre-training on the educational corpus and stimulate various skills with tool use by fine-tuning on designed system prompts and instructions. Currently, EduChat is available online as an open-source project, with its code, data, and model parameters available on platforms (e.g., GitHub https://github.com/icalk-nlp/EduChat, Hugging Face https://huggingface.co/ecnu-icalk ). We also prepare a demonstration of its capabilities online (https://vimeo.com/851004454). This initiative aims to promote research and applications of LLMs for intelligent education.

cs.CL↗