Approximate $k$ nearest neighbor (AKNN) search in high-dimensional space is a foundational problem in vector databases with widespread applications. Among the numerous AKNN indexes, Proximity Graph-based indexes achieve state-of-the-art search efficiency across various benchmarks. However, their extensive distance computations of high-dimensional vectors lead to slow construction and substantial memory overhead. The limited memory capacity often prevents building the entire index at once when handling large-scale datasets. A common practice is to build multiple sub-indexes separately. However, directly searching on these separated indexes severely compromises search efficiency, as queries cannot leverage cross-graph connections. Therefore, efficient graph index merging is crucial for multi-index searching. In this paper, we focus on efficient two-index merging and the merge order of multiple indexes for AKNN search. To achieve this, we propose a reverse neighbor sliding merge (RNSM) that exploits structural information to boost merging efficiency. We further investigate merge order selection (MOS) to reduce the merging cost by eliminating redundant merge operations. Experiments show that our approach yields up to a 5.48$\times$ speedup over existing index merge methods and 9.92$\times$ speedup over index reconstruction, while maintaining expected superior search performance. Moreover, our method scales efficiently to 100 million vectors with 50 partitions, maintaining consistent speedups.

Abrupt transitions ("tipping") in nonlinear dynamical systems are often accompanied by changes in the geometry of the attracting set, but quantifying such changes from partial and noisy observations in high-dimensional systems remains challenging. We address this problem with a sequential diagnostic framework, Data Assimilation-High dimensional Attractor's Structural Complexity (DA-HASC). First, this method reconstructs system's high-dimensional state using data assimilation from limited and noisy observations. Second, we quantify a structural complexity of the high-dimensional system dynamics from the reconstructed state by manifold learning. Third, we capture underlying changes in the system by splitting the reconstructed timeseries into sliding windows and analyzing the changes in the temporally local attractor's structural complexity. The structural information is provided as graph Laplacian and measured by Von Neumann entropy in this framework. We evaluate DA-HASC on both synthetic and real-world datasets and demonstrate that it can detect tipping under high-dimensionality and imperfect system knowledge. We further discuss how this framework behaves across different tipping mechanisms.

Nature plays a crucial role in human health and well-being, but little is known about how blind people experience and relate to it. We conducted a survey of nature relatedness with blind (N=20) and sighted (N=20) participants, along with in-depth interviews with 16 blind participants, to examine how blind people engage with nature and the factors shaping this engagement. Our survey results revealed lower levels of nature relatedness among blind participants compared to sighted peers. Our interview study further highlighted: 1) current practices and challenges of nature engagement, 2) attitudes and values that shape engagement, and 3) expectations for assistive technologies that support safe and meaningful engagement. We also provide design implications to guide future technologies that support nature engagement for blind people. Overall, our findings illustrate how blind people experience nature beyond vision and lay a foundation for technologies that support inclusive nature engagement.

Agentic Coding, powered by autonomous agents such as GitHub Copilot and Cursor, enables developers to generate code, tests, and pull requests from natural language instructions alone. While this accelerates implementation, it produces larger volumes of code per pull request, shifting the burden from implementers to reviewers. In practice, a notable portion of AI-generated code is eventually deleted during review, yet reviewers must still examine such code before deciding to remove it. No prior work has explored methods to help reviewers efficiently identify code that will be removed.In this paper, we propose a prediction model that identifies functions likely to be deleted during PR review. Our results show that functions deleted for different reasons exhibit distinct characteristics, and our model achieves an AUC of 87.1%. These findings suggest that predictive approaches can help reviewers prioritize their efforts on essential code.

We explore local risk-minimization, a quadratic hedging method for incomplete markets, in exponential additive models. The objectives are to derive explicit mathematical expressions and to conduct numerical experiments. While local risk-minimization is well studied for Lévy processes, little is known for the additive process case because, unlike Lévy processes, the Lévy measure for an additive process depends on time, which significantly complicates the mathematical framework. This paper shall provide a set of necessary conditions for deriving expressions for LRM strategies in exponential additive models, as integrability conditions on the Lévy measure, which allow us to confirm whether these conditions are satisfied for given concrete models. In the final section, we introduce the variance-gamma scaled self-decomposable process, a Sato process that generalizes the variance-gamma process, as a primary example, and perform numerical experiments.

Piecewise deterministic Markov process samplers are attractive alternatives to Metropolis--Hastings algorithms. A central design question is how to incorporate partial velocity refreshment to ensure ergodicity without injecting excessive noise. Forward Event-Chain Monte Carlo (FECMC) is a generalization of the Bouncy Particle Sampler (BPS) that addresses this issue through a stochastic reflection mechanism, thereby reducing reliance on global refreshment moves. Despite promising empirical performance, its theoretical efficiency remains largely unexplored. We develop a high-dimensional scaling analysis for standard Gaussian targets and prove that the negative log-density (or potential) process of FECMC converges to an Ornstein--Uhlenbeck diffusion, under the same scaling as BPS. We derive closed-form expressions for the limiting diffusion coefficients of both methods by analyzing their associated radial momentum processes and solving the corresponding Poisson equations. These expressions yield a sharp efficiency comparison: the diffusion coefficient of FECMC is strictly larger than that of optimally tuned BPS, and the optimum for FECMC is attained at zero global refreshment. Specifically, they imply an approximately eightfold increase in effective sample size per event over optimal BPS. Numerical experiments confirm the predicted diffusion coefficients and show that the resulting efficiency gains remain substantial for a range of non-Gaussian targets. Finally, as an application of these results, we propose an asymptotic variance estimator for Piecewise deterministic Markov processes that becomes increasingly efficient in high dimensions by extracting information from the velocity variable.

Reminiscence therapy (RT) is a common non-pharmacological intervention in dementia care. Recent technology-mediated interventions have largely focused on people with dementia through solutions that replace human facilitators with conversational agents. However, the relational work of facilitation is critical in the effectiveness of RT. Hence, we developed Rememo, a therapist-oriented tool that integrates Generative AI to support and enrich human facilitation in RT. Our tool aims to support the infrastructural and cultural challenges that therapists in Singapore face. In this research, we contribute the Rememo system as a therapist's tool for personalized RT developed through sociotechnically-aware research-through-design. Through studying this system in-situ, our research extends our understanding of human-AI collaboration for care work. We discuss the implications of designing AI-enabled systems that respect the relational dynamics in care contexts, and argue for a rethinking of synthetic imagery as a therapeutic support for memory rahter than a record of truth.

Friction losses in rough pipes are often predicted using semi-empirical correlations, such as the Colebrook-White equation (Colebrook,1939), which do not fully replicate Nikuradse's rough-pipe experiments (1950). This study derives scaling relations for the viscous and turbulent contributions to the streamwise pressure drop through an order-of-magnitude analysis of the Reynolds-averaged Navier-Stokes equations and the kinetic-energy transport equations. These relations impose constraints on the local sensitivity of the pressure drop to factors such as mean velocity, roughness, viscosity, and density through exponent envelopes and serve as a physical prior for symbolic regression. By combining Nikuradse's rough-pipe and smooth-pipe data of Zagarola and Smits (1998), we aim to derive compact correlations for the friction factor that fit experimental data while adhering to the derived constraints. A modified genetic programming engine (GPTIPS2) optimizes model structure and evaluates it based on fitness, complexity, and constraint violation. This method yields interpretable expressions that accurately reproduce friction factors across various roughness levels and Reynolds numbers, validated up to $Re \sim 10^7$.

Control of acoustic streaming can significantly impact fluid and particle transport in microfluidics. We report enhancement, suppression, and reversal of acoustic streaming inside a rectangular microchannel by controlling the fluid viscoelastic properties. Our study reveals that the streaming regimes depend on Deborah number ($De$) and viscous diffusion number ($Dv$), expressed in terms of a Streaming Coefficient ($C_s$). We find streaming is enhanced when $C_s>1$, suppressed for $0\leq C_s\leq1$, and reversed when $C_s<0$. We explain the regimes in terms of the interplay between the Reynolds and viscoelastic stresses that collectively drive fluid motion. Remarkably, we discover the role of viscoelastic shear waves in acoustic streaming transition characterized by the ratio of acoustic attenuation length and shear wavelength. We gain deeper insight into the streaming transition by examining energy dynamics in terms of the loss and storage moduli. Our study may find applications in acousto-microfluidics systems for particle handling and fluid pumping/mixing.

Coupled human-environment systems are increasingly being understood as complex adaptive systems (CAS), in which micro-level interactions between components lead to emergent behavior. Agent-based models (ABMs) hold great promise for environmental policy design by capturing such complex behavior, enabling a sophisticated understanding of potential interventions. One limitation, however, is that ABMs can be computationally costly to simulate, which hinders their use for policy optimization. To address this, we propose a new statistical framework that exploits machine learning techniques to accelerate policy optimization with costly ABMs. We first develop a statistical approach for sensitivity testing of the optimal policy, then leverage a reinforcement learning method for efficient policy optimization. We test this framework on the classic ``Sugarscape'' model, an ABM for resource harvesting. We show that our approach can quickly identify optimal and interpretable policies that improve upon baseline techniques, with insightful sensitivity and dynamic analyses that connect back to economic theory.

Multi-agent reinforcement learning (MARL) has made significant progress in recent years, but most algorithms still rely on a discrete-time Markov Decision Process (MDP) with fixed decision intervals. This formulation is often ill-suited for complex multi-agent dynamics, particularly in high-frequency or irregular time-interval settings, leading to degraded performance and motivating the development of continuous-time MARL (CT-MARL). Existing CT-MARL methods are mainly built on Hamilton-Jacobi-Bellman (HJB) equations. However, they rarely account for safety constraints such as collision penalties, since these introduce discontinuities that make HJB-based learning difficult. To address this challenge, we propose a continuous-time constrained MDP (CT-CMDP) formulation and a novel MARL framework that transforms discrete MDPs into CT-CMDPs via an epigraph-based reformulation. We then solve this by proposing a novel physics-informed neural network (PINN)-based actor-critic method that enables stable and efficient optimization in continuous time. We evaluate our approach on continuous-time safe multi-particle environments (MPE) and safe multi-agent MuJoCo benchmarks. Results demonstrate smoother value approximations, more stable training, and improved performance over safe MARL baselines, validating the effectiveness and robustness of our method.

Refining previous results, we establish a sharp asymptotic estimate on the expected graph distance between the origin and the terminal point of the trace of the first $n$ steps of the walk. A similar conclusion is drawn for the resistance metric.

Let $A$ be a unital $C^*$-algebra, and let $Σ^2_m A$ denote the $m$-torsioned quantum double suspension of $A$. For $q \in (0,1)$ and $n \geq 1$, we prove that the $C^*$-algebra corresponding to the quotient space $SO_q(2n+1)/SO_q(2n-1)$ is isomorphic to $Σ^{2(n-1)} \, Σ^2_2 \, Σ^{2(n-1)} C(\mathbb{T})$. It follows as a consequence that these spaces are independent of the deformation parameter $q$.

Harnessing dipolar spin environments as controllable quantum resources is a central challenge in solid-state quantum technologies. Here, we report the observation of a coherent mesoscopic spin state in a disordered ensemble of substitutional nitrogen (P1) centers in diamond. An iterative Hartmann-Hahn protocol transfers polarization from dense nitrogen-vacancy (NV) centers to a P1 ensemble, yielding a 740-fold enhancement over room-temperature thermal equilibrium as revealed by differential readout. The resulting mesoscopic P1 spin ensemble exhibits collective Rabi oscillations and long-lived spin-lock and Hahn-echo coherences. We identify a crossover in the saturation polarization arising from the competition between coherent driving and local disorder, providing a quantitative measure of the system's intrinsic disorder. These results establish a foundation for utilizing dark electron spin ensembles as robust resources for quantum sensing and quantum many-body simulation.

We extend the Abelian-Higgs model of superconductivity to incorporate higher-spin particles. Microscopically, these higher-spin states can be modeled as multi-electron clusters, such as spin-1 Copper pairs or quartets, existing alongside the standard Cooper pairs predicted by BCS theory. To account for these composites, we introduce vector and higher-rank tensor non-gauge fields into the Lagrangian, which serve as sources for higher-rank tensor gauge fields. In this work, we extend the particle spectrum by one rank (including the necessary auxiliary fields) and examine the resulting modifications to the fundamental phenomenological parameters of superconductivity, specifically the penetration depth and the correlation length.

We investigate elliptic and parabolic equations involving mixed local and nonlocal operators of the form $(-Δ)^s-Δ$, as well as their parabolic counterparts with both the Marchaud fractional time derivative and the classical first-order derivative. A major difficulty in this setting stems from the coexistence of operators with different nonlocal structures and incompatible scaling properties, which obstruct the direct use of classical sliding methods. To address this issue, we develop a refined sliding method suited to mixed local-nonlocal operators. As key technical ingredients, we establish new generalized weighted average inequalities, narrow region principles, and maximum principles in bounded and unbounded domains. These tools enable us to derive monotonicity and one-dimensional symmetry results for mixed elliptic equations in bounded domains, half-spaces, and the whole space, and to extend the analysis to parabolic equations with mixed time derivatives. As an application, we resolve the Gibbons' conjecture for a class of mixed fractional equations.

Personalized feedback plays an important role in self-regulated learning (SRL), helping students track progress and refine their strategies. However, current common solutions, such as text-based reports or learning analytics dashboards, often suffer from poor interpretability, monotonous presentation, and limited explainability. To overcome these challenges, we present StoryLensEdu, a narrative-driven multi-agent system that automatically generates intuitive, engaging, and interactive learning reports. StoryLensEdu integrates three agents: a Data Analyst that extracts data insights based on a learning objective centered structure, a Teacher that ensures educational relevance and offers actionable suggestions, and a Storyteller that organizes these insights using the Heroes Journey narrative framework. StoryLensEdu supports post-generation interactive question answering to improve explainability and user engagement. We conducted a formative study in a real high school and iteratively developed StoryLensEdu in collaboration with an e-learning team to inform our design. Evaluation with real users shows that StoryLensEdu enhances engagement and promotes a deeper understanding of the learning process.

Quantum communication holds the potential to revolutionize information transmission by enabling secure data exchange that exceeds the limits of classical systems. One of the key performance metrics in quantum information theory, namely the Holevo bound, quantifies the amount of classical information that can be transmitted reliably over a quantum channel. However, computing and optimizing the Holevo bound remains a challenging task due to its dependence on both the quantum input ensemble and the quantum channel. In order to maximize the Holevo bound, we propose a unified projected gradient ascent algorithm to optimize the quantum channel given a fixed input ensemble. We provide a detailed complexity analysis for the proposed algorithm. Simulation results demonstrate that the proposed quantum channel optimization yields higher Holevo bounds than input ensemble optimization.

Iterative algorithms are fundamental tools for approximating fixed-points of nonexpansive operators in real Hilbert spaces. Among them, Krasnosel'ski\uı--Mann iteration and Halpern iteration are two widely used schemes. In this work, we formalize the convergence of these two fixed-point algorithms in the interactive theorem prover Lean4 based on type dependent theory. To this end, weak convergence and topological properties in the infinite-dimensional real Hilbert space are formalized. Definition and properties of nonexpansive operators are also provided. As a useful tool in convex analysis, we then formalize the Fejér monotone sequence. Building on these foundations, we verify the convergence of both the iteration schemes. Our formalization provides reusable components for machine-checked convergence analysis of fixed-point iterations and theories of convex analysis in real Hilbert spaces. Our code is available at https://github.com/TTony2019/fixed-point-iterations-in-lean.

Characterizing the intensity, phase, and polarization of engineered light is fundamental to understanding and applying metasurfaces. However, existing characterization frameworks are hindered by several limitations, most notably their inability to account for the polarization of the field. Here, we report polarization sensitive coherent modulation imaging (PS-CMI), a light-weight but robust, high-resolution platform for the full-field characterization of metasurface-modulated light. By supplementing the orthogonal x- and y- complex amplitude components with an additional 45°-component, this approach calculates the retardance between two orthogonal polarization components while eliminating phase offsets, thereby enabling the subsequent recovery of the complete polarization state. We demonstrate the versatility of our method by characterizing light fields produced by a United States Air Force (USAF) target, two kinds of complex polarization field, and a metalens. This compact solution addresses a critical gap in metasurface metrology and is broadly applicable to other fields requiring the mapping of complex, polarized light distributions.

Accurately modeling long-term value (LTV) at the ranking stage of short-video recommendation remains challenging. While delayed feedback and extended engagement have been explored, fine-grained attribution and robust position normalization at billion-scale are still underdeveloped. We propose a practical ranking-stage LTV framework addressing three challenges: position bias, attribution ambiguity, and temporal limitations. (1) Position bias: We introduce a Position-aware Debias Quantile (PDQ) module that normalizes engagement via quantile-based distributions, enabling position-robust LTV estimation without architectural changes. (2) Attribution ambiguity: We propose a multi-dimensional attribution module that learns continuous attribution strengths across contextual, behavioral, and content signals, replacing static rules to capture nuanced inter-video influence. A customized hybrid loss with explicit noise filtering improves causal clarity. (3) Temporal limitations: We present a cross-temporal author modeling module that builds censoring-aware, day-level LTV targets to capture creator-driven re-engagement over longer horizons; the design is extensible to other dimensions (e.g., topics, styles). Offline studies and online A/B tests show significant improvements in LTV metrics and stable trade-offs with short-term objectives. Implemented as task augmentation within an existing ranking model, the framework supports efficient training and serving, and has been deployed at billion-scale in Taobao's production system, delivering sustained engagement gains while remaining compatible with industrial constraints.

Several factors may be responsible for disorder and frustration in a magnetic nanoparticle, including atomic vacancies on the surface and inside, impurity atoms, long-range magnetic exchange coupling, etc. We use Monte-Carlo simulations within the Heisenberg model to examine the role of randomly distributed atomic vacancies and long-range magnetic-exchange coupling on the temperature-dependent magnetic properties of ferromagnetic nanoparticles. In particular, we study the role of the second-neighbor antiferromagnetic exchange coupling and missing atoms inside the particle resulting in broken nearby bonds. We find that both factors may enhance the superparamagnetic behaviors of such particles.

We present a comprehensive experimental and theoretical investigation of the evaporation dynamics of freely levitated water droplets in an upward airstream under varying temperature and relative humidity conditions, using a custom-designed wind tunnel that replicates natural rainfall scenarios. A high-speed imaging system captures the temporal evolution of morphology, shape oscillations, and size reduction of the droplet undergoing evaporation. Our observations reveal that larger droplets exhibit persistent shape oscillations due to the interplay between inertia and surface tension in the presence of convective airflow, which significantly alters the evaporation rate compared to that of a stationary spherical droplet in quiescent air. To quantify the effects of air convection, complex morphology, and shape oscillations of the levitated droplet at different temperatures and humidity, we develop a modified evaporation model that extends the classical $d^2$-law. This model incorporates (i) a generalized Sherwood number that accounts for the variation in Reynolds number, Schmidt number, temperature, and relative humidity and (ii) a shape factor that captures the time-averaged surface area of oscillating droplets. The model is validated against experimental findings across a wide range of droplet sizes and environmental conditions, showing excellent agreement in predicting the temporal evolution of droplet diameter and total evaporation time. Furthermore, we construct a regime map showing the variation in the lifetime of the droplet in the temperature-humidity space. The present study establishes a framework that integrates convective transport and morphological deformation, offering new insights into the microphysics of raindrop evaporation.

As power systems accommodate higher shares of renewable generation, short-term power imbalances become more frequent and can manifest as pronounced voltage and frequency excursions under low-inertia conditions. E-STATCOMs (STATCOMs equipped with energy storage) offer a practical means to provide both voltage support and fast frequency assistance under grid-forming control. Among candidate implementations, double-star multilevel-converter (DS-MC)-based E-STATCOMs enable centralized energy-storage integration at the dc link, which improves thermal management and maintainability. Nevertheless, conventional dc-side power-based internal-energy regulation in DS-MCs can undesirably couple loss compensation to the energy-storage path, accelerating storage cycling and constraining operation when the storage is unavailable. This paper introduces a control strategy that assigns DS-MC total internal-energy regulation to the ac-side active-power path, while reserving dc-side storage power solely for frequency support. By decoupling internal-energy management from inertial-response provision, the proposed scheme enables flexible operation as either a STATCOM or an E-STATCOM according to storage availability and mitigates unnecessary storage cycling. The proposed strategy is verified through offline simulations and laboratory-scale experiments.

Generative modeling builds on and substantially advances the classical idea of simulating synthetic data from observed samples. This paper shows that this principle is not only natural but also theoretically well-founded for bootstrap inference: it yields statistically valid confidence intervals that apply simultaneously to both regular and irregular estimators, including settings in which Efron's bootstrap fails. In this sense, the generative modeling-based bootstrap can be viewed as a modern version of the smoothed bootstrap: it could mitigate the curse of dimensionality and remain effective in challenging regimes where estimators may lack root-$n$ consistency or a Gaussian limit.

Personalized image retouching aims to adapt retouching style of individual users from reference examples, but existing methods often require user-specific fine-tuning or fail to generalize effectively. To address these challenges, we introduce $\textbf{InstantRetouch}$, a general framework for personalized image retouching that instantly adapts to user retouching styles without any test-time fine-tuning. It employs an $\textit{asymmetric auto-encoder}$ to encode the retouching style from paired examples into a content disentangled latent representation that enables faithful transfer of the retouching style to new images. To adaptively apply the encoded retouching style to new images, we further propose $\textit{retrieval-augmented retouching}$ (RAR), which retrieves and aggregates style latents from reference pairs most similar in content to the query image. With these components, $\textbf{InstantRetouch}$ enables superior and generic content-aware retouching personalization across diverse scenarios, including single-reference, multi-reference, and mixed-style setups, while also generalizing out of the box to photorealistic style transfer.

Because the decathlon tests many facets of athleticism, including sprinting, throwing, jumping, and endurance, many consider it to be the ultimate test of athletic ability. On this view, estimating the maximal decathlon score and understanding what it would take to achieve that score provides insight into the upper limits of human athletic potential. To this end, we develop a Bayesian composition model for forecasting how individual athletes perform in each of the 10 decathlon events of time. Besides capturing potential non-linear temporal trends in performance, our model carefully captures the dependence between performance in an event and all preceding events. Using our model, we can simulate and evaluate the distribution of the maximal possible scores and identify profiles of athletes who could realistically attain scores approaching this limit.

Quantum cascade lasers (QCLs) are unipolar semiconductor lasers first demonstrated in 1994. Since then, they have played a central role in advancing mid-infrared and terahertz photonics, becoming among the most reliable light sources in these regions of the electromagnetic spectrum. Their importance is further reinforced by their ability to generate self-starting optical frequency combs, whose investigation is motivated both by fundamental physics and by a wide range of applications, including molecular spectroscopy and free-space optical communications. This Roadmap provides a unified overview of current advances and emerging directions in QCL research. The chapters are organized into three main sections: device design and technology; frequency combs and pulse formation; and applications of QCLs. Each chapter reviews the relevant background, summarizes the current state of the art, and identifies key challenges and future directions within its specific research area.

Recently, generating 3D assets with the control of condition images has achieved impressive quality. However, existing 3D generation methods are limited to handling a single control objective and lack the ability to utilize multiple images to independently control different regions of a 3D asset, which hinders their flexibility in applications. We propose Fuse3D, a novel method that enables generating 3D assets under the control of multiple images, allowing for the seamless fusion of multi-level regional controls from global views to intricate local details. First, we introduce a Multi-Condition Fusion Module to integrate the visual features from multiple image regions. Then, we propose a method to automatically align user-selected 2D image regions with their associated 3D regions based on semantic cues. Finally, to resolve control conflicts and enhance local control features from multi-condition images, we introduce a Local Attention Enhancement Strategy that flexibly balances region-specific feature fusion. Overall, we introduce the first method capable of controllable 3D asset generation from multiple condition images. The experimental results indicate that Fuse3D can flexibly fuse multiple 2D image regions into coherent 3D structures, resulting in high-quality 3D assets. Code and data for this paper are at https://jinnmnm.github.io/Fuse3d.github.io/.

Feedback from active galactic nuclei (AGN) is widely acknowledged to regulate the growth of massive galaxies, though its driving mechanisms are debated. Prevailing theories suggest that AGN-driven outflows are driven either by radiation pressure acting directly on the dusty interstellar medium (ISM) or by hot winds entraining cooler ISM gas, but the relative contribution of each mechanism remains uncertain. By combining optical emission line measurements with highly ionized UV emission lines, it is possible to constrain whether the pressure source applied to ionized clouds is primarily radiation or primarily hydrodynamic, and thus constrain the dominant driver. This study presents the first multi-object analysis of far-ultraviolet (FUV) spectra from galactic-scale AGN-driven outflows in obscured quasars, based on Cosmic Origins Spectrograph observations of five low-redshift targets. By comparing narrow-line region UV emission line ratios to theoretical models that vary the importance of the two pressure sources, we find three out of five targets fall within the radiation pressure-dominated regime. A fourth target exhibits intermediate emission-line ratios that suggest radiation pressure and pressure from a hot wind are both dynamically important. Finally, the lowest-luminosity object in our sample may have a dynamically important hot wind component, but non-detections prevent a clear conclusion in this case. These results suggest radiation pressure dominates circum-nuclear narrow-line region cloud dynamics, but pressure from a hot wind also plays a role in some cases. This is consistent with AGN feedback scenarios mediated by radiation pressure or a short-lived hot wind phase that dissipates after initially accelerating outflows.