Searcharxiv⌕ Search

arXiv subjects

Qi Wu

Publications and source records attributed to Qi Wu.

At least 19 recordsLinked to original sources

Study on the charmless decay modes of $Z_{cs}(3985)$

In this work, we investigate the charmless decay modes of the hidden-charm tetraquark state with strangeness, $Z_{cs}(3985)$, assuming it to be an $S$-wave hadronic molecule of $D_{s}^{*-}D^{0}$ and $D_{s}^{-}D^{*0}$ with quantum numbers $I(J^{P}) = \frac{1}{2}(1^{+})$. Using an effective Lagrangian approach, we calculate the decay amplitudes for $Z_{cs} \to \mathcal{PV}$ (where $\mathcal P$ and $\mathcal V$ denote light pseudoscalar and vector mesons, respectively) via triangle loop diagrams. We use the experimental upper limit on $Z_c(3900)\to πω$ to constrain the cutoff parameter for both the dipole and tripole form factors. Numerical results from the dipole form factor show that the branching ratios for $Z_{cs} \to πK^*$ can reach up to the order of $10^{-2}$, which is of the same order of magnitude as the hidden charm decay modes and may be detected in future measurements. The branching ratios from tripole form factor are not consistent with those from the dipole form factor, indicating a certain degree of model dependence in the present work. The relative ratios of the branching ratios for different decay channels are found to be almost independent of the cutoff parameter. However, some ratios are highly dependent on the choice of the dipole or tripole form factor. The branching ratio of $Z_{cs} \to ηK^{*}$ is found to be highly sensitive to the $η$-$η^\prime$ mixing angle, while that for $Z_{cs} \to η^\prime K^{*}$ is almost independent of the mixing angle, making it a clean channel for future experimental measurements.

hep-ph↗

P$^2$-MOPD: Accelerating Multi-Teacher On-Policy Distillation via Parameter Priors

Multi-teacher on-policy distillation consolidates expert capabilities by supervising student-generated trajectories, whereas task arithmetic offers a training-free alternative through linear parameter combinations. We view the linear combination of student and expert parameters as a single differentiable model and propose P$^2$-MOPD, which jointly optimizes the mixing coefficients and the student weights under the same on-policy distillation objective. Unlike task arithmetic, which stays within the span of the frozen student and expert parameters for any choice of merging coefficients, P$^2$-MOPD keeps the student trainable, so the experts act as parameter priors without confining it to this span. Across four domains at 1.7B and 4B scales, P$^2$-MOPD achieves faster and more balanced recovery than MOPD from the same initialization, surpassing its final average recovery in half the training steps. At the end of training, our 1.7B model reaches 94.7% average recovery, versus 80.9% for MOPD.

cs.LG↗

On the Turán number for the Cartesian product of an odd cycle and a tree

Let $T$ be a fixed tree on at least two vertices. We determine the Turán number and all extremal graphs for the Cartesian product of an odd cycle $C_{2m+1}$ and $T$, for every fixed integer $m\ge1$ and all sufficiently large $n$, in terms of its decomposition family. Every extremal graph is the join of an independent set and an extremal graph for that family. The proof combines the Erdős--Sós theorem with weak stability and progressive induction.

math.CO↗

Radiative decays of the predicted P-wave $D\bar{D}^*$ molecular state: The $0^{-+}$ case

In the present work, we study the radiative decays of the $P$-wave $D\bar{D}^*$ molecular state $G_0$ with $I^G(J^{PC})=0^+(0^{-+})$ using a phenomenological Lagrangian approach. The molecular coupling is fixed by the compositeness condition with a Gaussian form factor. The partial widths for $G_0\to γγ$, $γJ/ψ$, $γψ(2S)$, $γh_c$, $γρ$, and $γω$ are evaluated. Our results show that the two-photon and $γh_c$ channels are highly suppressed, while $γJ/ψ$, $γψ(2S)$, and $γρ$ are of the order of a few keV, and $γω$ reaches several tens of keV, making it the dominant radiative mode. The ratios $Γ(G_0 \to γψ(2S))/Γ(G_0 \to γJ/ψ)$ and $Γ(G_0 \to γω)/Γ(G_0 \to γρ)$ are predicted to be $1-2.6$ and $3.4-5.4$, respectively, with mild parameter dependence. We also compare with hadronic decays and find that the total radiative decay width is lower than the hadronic decay widths by about one order of magnitude. We propose searching for $G_0$ in the sequential decay $Y(4230)\to γG_0 \to γωJ/ψ$, which is accessible to further experimental measurements by the BESIII, Belle II and LHCb Collaborations.

hep-ph↗

TerraVis: Towards Evaluation of World-Grounded Visual Consistency in Text-to-Image Generation via MLLM Workflows

Recent text-to-image models have made substantial progress in photorealism, aesthetics, and text-image alignment. Yet visually appealing images can still violate real-world plausibility, exhibiting malformed object structures, impossible anatomy, physically implausible interactions, or inconsistent spatial relationships. Such failures are not well captured by existing fidelity, aesthetics, preference, or alignment metrics. To address this gap, we introduce TerraVis, a framework for evaluating world-grounded visual consistency in generated images. TerraVis defines a structured taxonomy of world-consistency violations spanning object-, interaction-, and scene-level failures, and employs a multi-stage evaluation framework to identify and quantify them. Given an image, TerraVis first uses an MLLM to assess its eligibility for evaluation, then detects violations across 18 taxonomy-defined types and classifies them as minor or major to derive an overall world-consistency score. Across diverse open-source and proprietary text-to-image models on two widely used benchmarks, TerraVis achieves the strongest correlation with human judgments of world consistency among existing metrics. Our benchmark results further show that models that achieve strong performance on conventional metrics can still exhibit substantial world-consistency failures. These findings highlight world consistency as a complementary evaluation dimension and demonstrate that TerraVis enables systematic quantification, diagnosis, and comparison of such failures. Our code is publicly available at https://github.com/ShyFoo/TerraVis.

cs.CV↗

Structure over Reasoning: Unified Zero-Shot Navigation with a Small Vision-Language Model

Vision-language navigation is moving into photorealistic, physically simulated scenes, where the leading zero-shot agents put a vision-language model in the control loop to choose every move and stop. Robots favour small models, whose abilities are uneven: Qwen3-VL-4B can judge whether a single view looks like the goal, but its route reasoning is at chance, its distance estimates nearly constant, its stop votes unreliable. Our agent therefore asks it two perceptual questions per step, which direction looks like the goal and whether the view justifies stopping, plus one gated frontier query on route-bearing instructions, and treats the answers as evidence, not actions: a route-blind semantic field over a bird's-eye map accumulates the directional scores through a tempered update, a skeleton graph supplies frontiers and geodesic distances, and a two-stage terminator turns stop votes into termini through a trail-consistency gate and a budget-forced return. With every baseline re-run in one installation at one budget, the agent exceeds all four zero-shot baselines of the VLNVerse subset in mean success on both granularities, though the paired interval against the strongest coarse-grained baseline includes zero. Termination is the dominant lever in ablation, thirteen attempts to give the model more of the navigation establish no reliable gain, and the agent runs on a Unitree Go2 quadruped with off-board inference. Code, per-episode logs and a robot video will be released at https://anonymous.4open.science/r/structure-over-reasoning-PAGE.

cs.RO↗

Automating the Design of Embodied Agent Architectures

Embodied agents are typically built as hand-designed compositions of perception, memory, planning, and action modules. This modularity exposes a large architectural design space, but current systems still rely on researcher intuition to choose where information is stored, how observations are processed, and how model calls are connected. Agent Architecture Search (AAS) automates such design for text-domain agents, but has not been systematically evaluated on perceptual embodied agents. We study this transfer. To make embodied architectures searchable, we introduce AgentCanvas on the executor side, representing embodied agents as editable typed graphs with simulator execution and episode-level logs. On the optimizer side, we lift AAS from LLM-call loops to tool-using coding-agent sessions, adapting ADAS and AFlow and introducing Knowledge Distill Loop (KDLoop). Across four embodied executors (MapGPT, SmartWay, ExploreEQA, and VoxPoser), search discovers improvements that persist on episodes never queried during search. KDLoop and AFlow independently discover elevation-tolerant observation processing. KDLoop's implementation also improves success rates over the baseline when transferred to real robots. These improvements are not uniform: some gains remain uncertain or reverse on held-out episodes, and an apparent gain exploits evaluator-side information. The experiments further show how rollout noise, local edit basins, and incomplete use of episode-level evidence constrain search. Together, these results establish the potential of automated architecture search for embodied agents and characterize the challenges of applying it in simulated and real-world environments. Code is available at https://github.com/jianzhou0420/AgentCanvas.

cs.RO↗

Embodied Agents Take Control: Minimal-Interface Zero-Shot Agents Rival Industrial-Scale Policies in Vision-and-Language Navigation

Autonomous embodied agents must sustain a long decision-making loop of perceiving, acting, verifying, and self-correcting. In trained end-to-end policies and hand-designed workflows, the interaction structure remains externally specified. We study how much of this loop a general-purpose model can direct through a minimal interface. We term this organization agentic embodied control: the reasoning model directly steers every action, keeping reasoning and control aligned. We evaluate fourteen frontier multimodal models as provided by their vendors, with monocular RGB and primitive actions across six navigation and embodied question-answering benchmarks. On 100-episode subsets, the best agents exceed published scores and approach or surpass our human tester. Who directs the loop matters: SmartWay and O2C-Nav using the same strong models achieve substantially lower success rates on R2R-CE. The lightweight mini-swe-agent harness performs comparably to vendor harnesses. Stronger models benefit from optional rather than imposed waypoints; adding SLAM raises R2R-CE success to 88%. Physical trials expose limitations in body awareness, execution feedback, and persistent spatial understanding.

cs.RO↗

Triangle-free Rank--Ramsey graphs from ordered incidence lifts

For a graph $G$ with adjacency matrix $A_G$ and a coefficient field $\mathbb K$, put $ρ_{\mathbb K}(G)=\operatorname{rank}_{\mathbb K}(A_G+I)$. We answer the triangle-free Rank--Ramsey existence question of Beniamini, Linial and Shraibman. For every order $n$, we construct a triangle-free graph $H_n$ such that, simultaneously over all coefficient fields, $$ ρ_{\mathbb K}(H_n)=\bigl(4(2/3)^{3/4}+o(1)\bigr)n^{3/4}, $$ with an error uniform in the field. The graphs can be chosen Hamiltonian for all sufficiently large $n$. Our method combines ordered layers of point--line incidence flags with a common low-dimensional endpoint space. We determine the exact real rank, independence number and chromatic number of the affine lifts. For all triangle-free complete incidence lifts, we obtain a sharp two-term lower bound for complete incidence lifts, with sharpness witnessed by projective-plane incidence graphs over suitable coefficient fields. We also establish a third-order estimate and stability near equality. On the same explicit graphs, both the Lovász number and minimum semidefinite rank exceed the prescribed Boolean-matrix rank by polynomial factors. Thus neither parameter has a universal constant-factor upper bound in terms of complement rank. We determine Shannon capacity in the short-layer range.

math.CO↗

$ψ(3S)$ and $η_c(3S)$ production in the $B^+$ decay processes

In the present work, we investigate the production of psi(4040) in the B+ -> K+ psi(4040) process. Our estimations indicate that the unexpectedly large branching fraction could be well reproduced by taking both the short-range perturbative and long-range non-perturbative QCD contributions into account, where the long-range non-perturbative part plays an important role. Considering the structural similarity of eta_c(3S) and psi(3S), we apply the same model parameters to investigate the production of eta_c(3S) in the B+ decay process, and the branching fraction is predicted to be (0.81 +0.31 -0.29) x 10^-3. Based on the estimations in the present work and the relevant experimental measurements, we suggest that the experimentalists search for eta_c(3S) in the J/psi omega invariant mass distributions of B+ -> J/psi omega K+ process at Belle II and LHCb Collaborations.

hep-ph↗

Does Peer Observation Help? Vision-Sharing Collaboration for Vision-Language Navigation

Vision-Language Navigation (VLN) systems are fundamentally constrained by partial observability, as an agent can only accumulate knowledge from locations it has personally visited. As multiple robots increasingly coexist in shared environments, a natural question arises: can agents navigating the same space benefit from each other's observations? In this work, we introduce Co-VLN, a minimalist, model-agnostic framework for systematically investigating whether and how peer observations from concurrently navigating agents can benefit VLN. When independently navigating agents identify common traversed locations, they exchange structured perceptual memory, effectively expanding each agent's receptive field at no additional exploration cost. We validate our framework on the R2R benchmark under two representative paradigms (the learning-based DUET and the zero-shot MapGPT), and conduct extensive analytical experiments to systematically reveal the underlying dynamics of peer observation sharing in VLN. Results demonstrate that vision-sharing enabled model yields substantial performance improvements across both paradigms, establishing a strong foundation for future research in collaborative embodied navigation.

cs.CV↗

Optical Coherent Combining in Multi-Aperture Free-Space Receivers through Atmospheric Turbulence via Optically Computed Alignment Gradients

Multi-aperture reception converts spatial diversity into a coherent-combining gain, but only if the relative phase and polarization of the collected field replicas are aligned before they are summed. In a satellite-to-ground downlink the atmosphere perturbs amplitude, phase and polarization, so the alignment must track a fluctuating channel rather than a fixed offset. This paper proposes a new optical-domain combining scheme in which the alignment information is computed by light itself: the ascent directions that maximize the combined output power are formed by optical interference and read out directly as differential photocurrents of balanced detection, so no estimate of the channel, no deliberate perturbation, and no per-branch digital processing is needed to obtain them. Every branch is aligned against one shared reference - a passive tap of the combined field itself - so no independent laser or optical frequency lock is required, and the update directions of all branches are available simultaneously in a single averaging window without frequency, code, or time multiplexing. The data path needs a single dual-polarization coherent receiver regardless of the aperture count, and the readout is built from elementary parts only - a 3 dB directional coupler, static phase shifters, a Faraday rotator, and balanced photodiodes - so no 90-degree optical hybrid, mixer, or polarization-diverse hybrid is required. The phase and polarization loops are fixed-gain and division-free, and each branch carries one phase modulator and one polarization controller with no amplitude actuator. A closed-form combining-gain and loss budget is derived, the assumptions and limitations of the scheme are stated explicitly, and experimental validation remains open.

physics.optics↗

KinoGen: Customizable Humanoid Loco-Manipulation References via Kinematically Grounded Video Generation

We present KinoGen, a framework that synthesizes customizable humanoid loco-manipulation references through kinematically grounded video generation. KinoGen conditions a pretrained video generator on calibrated renders of the target robot and object in a simulated task instance and recovers the reference by fitting the same models to the generated two-view video, so that the reference lies in the robot configuration space by construction without a human body proxy or retargeting. Body, hand, and object priors initialize the reconstruction, and direct optimization against cross-view evidence refines it. Since an edited keyframe is simply another rendered state, a user can prescribe a goal object pose, regenerate the video around the edited keyframe, and obtain a reference that satisfies the pose exactly. Across eight loco-manipulation tasks, KinoGen yields lower penetration and foot skating than video-based baselines and preserves over 90% of annotated contacts, and its customized references satisfy every prescribed target with substantially less penetration than direct interpolation. Tracking policies trained on the synthesized references are deployed on a Unitree G1 humanoid with dexterous hands. Project webpage: https://kinogen-humanoid.github.io/.

cs.RO↗

NavGPT-3: Harnessing Context in a Hierarchical Navigation Runtime

Language models trained with long-horizon agentic reinforcement learning can generalize knowledge through reasoning, express precise actions, and pursue goals over many steps, raising the ceiling on what an embodied agent can understand and decide. Physical interaction, however, remains the domain of action policies, which provide dense, low-latency control. We present NavGPT-3, a harness that connects the two models, with an OS-like runtime built above it: reasoning, acting, and monitoring run as threads with their own context, tools, and permissions, while the runtime schedules them and decides which thread controls the robot's motion, so that the robot can react to sudden real-world events through interruption and thread switching. Beneath it, our action policy NavGPT VLA, trained on 19.28M examples, allocates visual tokens using codec allocation, in proportion to scene change; its 8B model alone reaches 76.94 SR on R2R-CE and leads RxR-CE with 82.39 SR. With the complete harness, NavGPT-3 sets the state of the art on R2R-CE (81.51 SR) and, for the first time, brings an autonomous agent to human level: on RxR-CE it matches human followers in success (90.43 vs. 90.4 SR) and path fidelity (78.47 vs. 77.7 nDTW) at 1 min 22 s per episode, versus roughly 3 min for a human. We comprehensively ablate the harness design and the interaction between the two models, showing how tools and the action policy shape the path from language-model reasoning to physical control: when NavGPT VLA executes the route, the reasoning loop shortens and the system's minimum reaction time falls from 3-19 s per language-model decision to 0.5-1 s per action-policy step (1-2 Hz). These results show that designing this embodied interface is central to connecting frontier language-model intelligence with low-level physical control. We will release all models, code, and evaluation records.

cs.RO↗

Data-Centric Recursive Self-Improvement for Navigation via Policy-Synchronized Language-Action Dual Bootstrapping

Vision-language navigation requires agents to follow natural-language instructions to reach target locations, making high-quality language supervision essential. Since human annotation is costly, prior work has explored scalable data synthesis methods. However, existing methods typically either use separate models for navigation and instruction generation or couple the two tasks through task-specific components; most also annotate the full trajectory corpus with a fixed model before updating, which can cause generated supervision to lag behind the evolving policy. Therefore, we introduce RSI-Nav, a data-centric recursive self-improvement framework for navigation via policy-synchronized language-action dual bootstrapping, which unifies vision-language-to-action instruction following and vision-action-to-language instruction generation within a shared model. Specifically, after supervised fine-tuning establishes fundamental capabilities, RSI-Nav progressively processes unlabeled trajectories in small shards. The current model generates instructions for these trajectories, executes them to evaluate goal completion and path efficiency, retains qualified samples, and immediately updates itself before processing the next shard. This shard-wise update keeps newly generated supervision aligned with the evolving policy. We also incorporate spatial understanding data, since stronger spatial reasoning benefits navigation. Extensive experiments show that RSI-Nav reaches 68.6% SR on R2R-CE and 72.4% SR on RxR-CE, while providing stronger instruction-generation utility and modestly improved spatial understanding after recursive evolution.

cs.CV↗

Long-Distance Contributions to Light Vector Mesons and Radiative Decays of $η_{c}$

In the present work, we investigate the long-distance contributions to light hadrons and radiative decays of $η_c$, where the long-distance contribution is phenomenologically described by the meson loops. Our estimations indicate that the branching fractions of $η_c \to \mathcal{VV},\ \mathcal{V}=(ρ,ω, ϕ)$ can be roughly reproduced in the range $1.4<α<1.6$. While the parameter range for $η_c \to γ\mathcal{V}$ is determined by the similarity of $η_c \to γ\mathcal{V}/\mathcal{VV}$ and $χ_{c1} \to γ\mathcal{V}/\mathcal{VV}$. In this parameter range, our estimations indicate that the branching fractions for the processes $η_c \to γρ^0$ and $η_c \to γω$ are of order $10^{-4}$, while the one for the process $η_c \to γϕ$ is of order $10^{-5}$. The estimations in the present work are expected to be accessible by future experiments at BESIII and Belle II.

hep-ph↗

Pressure-induced Lifshitz and quantum phase transitions in electron-doped cuprate superconductor

We report the first observations of a pressure-induced Lifshitz transition coupled with a quantum phase transition in electron-doped cuprate superconductor Pr0.87LaCe0.13CuO4 (PLCCO), by combining high-pressure electrical resistance, Hall coefficient and synchrotron X-ray diffraction (XRD) measurements at low temperatures. Our low-temperature Hall coefficient measurements reveal that the Hall coefficient decreases continuously and reaches zero at about 10 GPa (critical pressure of Pc1). Upon further compression beyond Pc1, Hall coefficient unexpectedly changes its sign from negative to positive, signaling a reconstruction of the Fermi surface from electron-dominated to hole-dominated topology. Concurrently, the superconducting transition temperature (Tc) exhibits a monotonic suppression, vanishing completely at 17.6 GPa (critical pressure of Pc2), where the system enters a non-superconducting metallic state. Our low temperature XRD measurements unequivocally demonstrate the absence of any structural phase transition across Pc1 and Pc2. Therefore, the sign change in RH at Pc1 is associated with a Lifshitz transition, which is never found in the compressed bulk electron- or hole-doped cuprate superconductors. Moreover, the quantum phase transition observed at Pc2 contrasts sharply with known high-pressure behavior of hole-doped cuprates, uncovering a fundamental difference on how pressure tunes the ground states of electron- versus hole-doped systems. These findings provide crucial insights into the different pressure responses on the interplay among Fermi surface topology, electronic correlations, and superconductivity between these two kinds of cuprate superconductors.

cond-mat.supr-con↗

Strict spectral supersaturation for cliques: extremal graphs and sharp thresholds

For every fixed $r\ge3$ and all sufficiently large $n$, we determine the largest adjacency spectral radius of an $n$-vertex graph with fewer than $q c_r(n)$ copies of $K_{r+1}$, where $1\le q<n/r$. Here $c_r(n)$ is the number of copies created by adding one edge to a largest part of the Turán graph. We also determine all extremal graphs. In most cases the extremal graph is obtained from an almost balanced complete multipartite graph by adding a star in one part. Two small values of $q$ require separate constructions, and an additional transition occurs when $r=3$ and $n\equiv2\pmod3$. Under the non-strict constraint, the unique extremal graph is obtained by adding a $q$-edge star to a largest part of the Turán graph. We determine the difference between the strict and non-strict values and prove that the sharp matching threshold is $\sqrt2(r-1)/r$. The proof treats separately the ranges $q=o(n)$, $q/m\toγ\in(0,1)$, and $q/m\to1$.

math.CO↗