Showing 1–50 of 1,008 results for author: Lin, G

FAST-DeepONet: Factor-Augmented Branch Representations for High-Dimensional PDE Inputs in the Small-Sample Regime

Authors: Jiyong Kwon, Bongseok Kim, Guang Lin

Abstract: Deep operator networks can become statistically unstable when partial differential equation inputs are observed at thousands of strongly correlated sensors but only a small number of operator samples is available. We introduce FAST-DeepONet, a branch representation combining a fixed spectral path with a regularized projection of the orthogonal residual, in which the directional penalty acts on the… ▽ More Deep operator networks can become statistically unstable when partial differential equation inputs are observed at thousands of strongly correlated sensors but only a small number of operator samples is available. We introduce FAST-DeepONet, a branch representation combining a fixed spectral path with a regularized projection of the orthogonal residual, in which the directional penalty acts on the effective residual map after each of its rows is normalized. On Navier--Stokes flow a plain DeepONet degrades from $0.0394$ to $0.1556$ mean relative $L_2$ error as the branch grows from $129$ to $8193$ coordinates, while FAST-DeepONet stays near $0.04$, so the sensor grid can be refined without a statistical penalty. Across independent test sets for Navier--Stokes flow, Darcy flow, and signed terminal wavefield prediction it lowers mean relative $L_2$ error by $4.7\%$ to $37.0\%$ with three to seven times fewer trainable parameters. A spectral-only branch sharing the same basis separates the two paths: the fixed spectral path carries the improvement on Navier--Stokes and Darcy, while terminal wave prediction requires the residual path together with its directional penalty. FAST-DeepONet targets coordinate-query architectures and trains on solution values alone. △ Less

Submitted 15 August, 2026; originally announced August 2026.

A Unified Mamba--MoE Surrogate for Closed-Loop Simulation and Measurement-Window Forecasting of Inverter Transients

Authors: Haoguang Wang, Huy Hoang Le, Akhila Kandivalasa, Christian Moya, Marcos Netto, Guang Lin

Abstract: This paper proposes a Mamba surrogate model with mixture-of-experts (MoE) routing to represent the transient dynamics of inverter-based resources. A Mamba surrogate model is a predictive machine learning model built on the Mamba architecture. MoE routing uses a router network to assign data-dependent weights to specialized subnetworks (experts). The resulting Mamba--MoE surrogate can perform two t… ▽ More This paper proposes a Mamba surrogate model with mixture-of-experts (MoE) routing to represent the transient dynamics of inverter-based resources. A Mamba surrogate model is a predictive machine learning model built on the Mamba architecture. MoE routing uses a router network to assign data-dependent weights to specialized subnetworks (experts). The resulting Mamba--MoE surrogate can perform two tasks: (i) closed-loop simulation and (ii) measurement-window forecasting of inverter transients. A single Mamba backbone with task conditioning and expert routing serves both tasks, replacing two separate specialists. Task-matched objectives fit each prediction form, and an adaptive conformal layer provides prediction intervals for both tasks. For the considered grid-following inverter, the unified surrogate model remains in the same low-error regime as a Mamba specialist pair while using 13% fewer parameters. The prediction intervals achieve 94--96% empirical mean marginal coverage across the two tasks. For transient dynamics---that is, beyond the vicinity of an equilibrium point---our surrogate model with MoE routing yields lower errors across all outputs in both tasks compared to a shared Mamba backbone without expert routing. A controller hardware-in-the-loop simulation validates our results and shows that adapting only the shared output head with limited measured data reduces held-out forecasting error. △ Less

Submitted 15 August, 2026; originally announced August 2026.

RouteTS: Frequency-Time Routing for Time Series Forecasting

Authors: Gaofeng Lin, Lei Duan

Abstract: Real-world time series inherently intertwine global periodic structures with localized non-stationary variations. Existing approaches process these heterogeneous dynamics within a single computational domain, incurring fundamental limitations: time-domain models suffer from periodic misalignment over long horizons, while frequency-domain models over-smooth transient spikes. We argue that the optim… ▽ More Real-world time series inherently intertwine global periodic structures with localized non-stationary variations. Existing approaches process these heterogeneous dynamics within a single computational domain, incurring fundamental limitations: time-domain models suffer from periodic misalignment over long horizons, while frequency-domain models over-smooth transient spikes. We argue that the optimal computational domain is not a property of the model, but of the data itself. Based on this principle, we propose RouteTS, a unified forecasting framework that partitions the frequency spectrum via amplitude routing and delegates components to their mathematically optimal domains. Dominant frequencies are processed by a complex-valued linear predictor in the frequency domain to preserve periodic structure, while residual spectral energy is reverted to the time domain and modeled by a lightweight MLP for local variations. Extensive experiments demonstrate that RouteTS achieves competitive prediction accuracy across diverse real-world datasets, with routing decisions guided by the underlying spectral signature. Furthermore, the lightweight design of RouteTS provides significant computational efficiency advantages, offering a principled solution to the longstanding dilemma between global periodicity and local transience. △ Less

Submitted 4 August, 2026; originally announced August 2026.

RGB-D Video Generation for Improving Human-to-Robot Object Handover Prediction

Authors: Tianyu Sun, Zhoujie Fu, Zihui Gao, Bang Zhang, Guosheng Lin

Abstract: Human-to-robot (H2R) object handover is a fundamental capability for human-robot collaboration, yet progress is hindered by the scarcity of large-scale, human-centric datasets and the significant sim-to-real gap. To address these challenges, we introduce Hand2Bot, an RGB-D video dataset that provides rich contextual information such as body posture and facial expressions, specifically collected fo… ▽ More Human-to-robot (H2R) object handover is a fundamental capability for human-robot collaboration, yet progress is hindered by the scarcity of large-scale, human-centric datasets and the significant sim-to-real gap. To address these challenges, we introduce Hand2Bot, an RGB-D video dataset that provides rich contextual information such as body posture and facial expressions, specifically collected for handover scenarios with real-world noise patterns. We further propose PassGen, a generative pipeline that leverages stable video diffusion and an Intention-Aware Temporal Face Encoder to synthesize realistic handover sequences while ensuring hand-object consistency. To bridge the sim-to-real gap, we implement a morphology-based depth editing strategy that replicates realistic sensor noise found in physical depth maps. Experimental evaluations demonstrate that our framework achieves high intention identification accuracy and low false trigger rates in both ablation studies and real-world deployment on a physical robot platform. Our results confirm that training on PassGen allows for robust zero-shot transfer and earlier intention anticipation compared to traditional hand-centric baselines, effectively enabling socially aware robotic behavior in shared workspaces. △ Less

Submitted 13 August, 2026; originally announced August 2026.

Symbolic Machine Learning for Vapor-Liquid Equilibrium Prediction in Cx-N2 Binary Mixtures

Authors: Bongseok Kim, Suman Chakraborty, Gary Huang, Mehek Mathur, Guang Lin, Li Qiao

Abstract: Accurate prediction of vapor--liquid equilibrium (VLE) for hydrocarbon-nitrogen mixtures remains challenging for cubic equations of state, particularly across broad ranges of composition and hydrocarbon chain length. While deep learning models can provide accurate predictions, they often lack interpretability and explicit analytical expressions. In this work, we propose a symbolic machine learning… ▽ More Accurate prediction of vapor--liquid equilibrium (VLE) for hydrocarbon-nitrogen mixtures remains challenging for cubic equations of state, particularly across broad ranges of composition and hydrocarbon chain length. While deep learning models can provide accurate predictions, they often lack interpretability and explicit analytical expressions. In this work, we propose a symbolic machine learning approach to discover interpretable symbolic corrections to Peng-Robinson equation-of-state (PR-EOS) predictions from experimental data. The proposed approach adopts a two-level strategy: symbolic expressions are first identified for individual hydrocarbon systems, after which their coefficients are represented as functions of carbon number to enable accurate prediction across different hydrocarbon systems. The results demonstrate significantly improved prediction accuracy over the original PR-EOS across all hydrocarbon-nitrogen systems. Overall, the proposed approach provides an interpretable symbolic correction framework for improving PR-EOS predictions of hydrocarbon-nitrogen VLE. △ Less

Submitted 6 August, 2026; originally announced August 2026.

Bridging Event Streams and DiT: Event-Guided Video Frame Interpolation

Authors: Guixu Lin, Yuyang Yu, Xiang Ji, Linyao Chen, Zhengwei Yin, Mengshun Hu, Mingdeng Cao, Shengfeng He, Yinqiang Zheng

Abstract: Latent diffusion models have recently advanced video frame interpolation by synthesizing intermediate frames between input images. However, handling large temporal gaps and complex motion remains challenging, often resulting in motion blur, structural distortions, and temporal inconsistencies. Event cameras provide high-temporal-resolution motion cues that are well suited for bridging these gaps a… ▽ More Latent diffusion models have recently advanced video frame interpolation by synthesizing intermediate frames between input images. However, handling large temporal gaps and complex motion remains challenging, often resulting in motion blur, structural distortions, and temporal inconsistencies. Event cameras provide high-temporal-resolution motion cues that are well suited for bridging these gaps and improving interpolation quality. To exploit this advantage without training an event-assisted model from scratch, we propose an adapter-based framework that incorporates event-derived cues into a pre-trained image-to-video diffusion model with minimal architectural changes. Specifically, our method leverages Image Warped Events (IWEs) and bidirectional sparse optical flow to provide spatially and temporally aligned guidance during generation. By injecting these event-guided structural and motion cues into the diffusion process, our approach reduces interpolation artifacts and improves both reconstruction fidelity and temporal coherence. Experimental results on real and synthetic benchmarks show that our method consistently outperforms existing state-of-the-art approaches. The project page is at https://joseph-lin-tech.github.io/BridgeEventDiT-VFI/. △ Less

Submitted 11 August, 2026; v1 submitted 11 August, 2026; originally announced August 2026.

Comments: https://joseph-lin-tech.github.io/BridgeEventDiT-VFI/

Deciding When to Switch: E-Processes for Adaptive Minimax Training for Generative Adversarial Nets

Authors: Hyunjoo Kim, Sicheng Wu, Agastya Venkatraman, Guang Lin, Sehwan Kim

Abstract: Modern data science increasingly gives rise to hypothesis-testing problems that are not naturally formulated in terms of parameters within prespecified statistical models. One important example is the dynamic evaluation of optimization algorithms, where decisions must be made during training about whether further updates remain beneficial or the algorithm should switch to a different phase. This i… ▽ More Modern data science increasingly gives rise to hypothesis-testing problems that are not naturally formulated in terms of parameters within prespecified statistical models. One important example is the dynamic evaluation of optimization algorithms, where decisions must be made during training about whether further updates remain beneficial or the algorithm should switch to a different phase. This issue is particularly relevant in stochastic min-max optimization. Generative adversarial networks (GANs) provide a canonical example, as their training requires repeated decisions about when to switch between discriminator and generator updates, yet existing methods typically rely on fixed update ratios or heuristic criteria. We formulate this switching problem as sequential hypothesis testing and develop an e-process-based adaptive training procedure. During discriminator updates, one e-process tests the null that the discriminator-induced separation between the empirical data distribution and the generator law remains below a target level. During generator updates, with the discriminator fixed, a second e-process tests the reverse null that this separation remains above a refresh level. Conditional on the observed training sample, we prove that fresh empirical indices and latent draws yield conditional e-values that can be accumulated into e-processes, providing anytime-valid Type I error control under adaptive model updates and data-dependent switching. Across multimodal synthetic distributions and image benchmark datasets, the proposed method matches or outperforms the best fixed-ratio baselines under several widely used GAN objectives. △ Less

Submitted 10 August, 2026; originally announced August 2026.

SheafIQ: Sheaf-Theoretic Information Quantification of Vector Fields on Geometric Graphs

Authors: Cong Shen, Guancen Lin, Chuan-Shen Hu

Abstract: Vector fields on graph structures naturally arise in diverse biological and engineered systems, where vector-valued states are defined on the nodes and evolve through the network interactions. Existing methods primarily characterize either the graph topology or individual signals, but generally do not quantify how local interactions among node-associated vectors are organized across the graph. To… ▽ More Vector fields on graph structures naturally arise in diverse biological and engineered systems, where vector-valued states are defined on the nodes and evolve through the network interactions. Existing methods primarily characterize either the graph topology or individual signals, but generally do not quantify how local interactions among node-associated vectors are organized across the graph. To address this limitation, a sheaf-theoretic framework, termed SheafIQ, is proposed to represent neighboring vectors in a common edge-associated coordinate system, map local incompatibilities to a residual energy distribution, and quantify its global organization through entropy. Across proteins, functional brain networks, urban traffic systems, and power grids, SheafIQ consistently reveals complementary organizational information beyond conventional graph- and signal-based descriptors. More broadly, it establishes a unified information-theoretic framework for quantifying the organization of vector-valued states on geometric graphs, extending network analysis beyond graph topology alone. △ Less

Submitted 9 August, 2026; originally announced August 2026.

GraspMeanFlow: SE(3)-Equivariant MeanFlow for Few-Step 6-DoF Grasp Generation

Authors: Jiyong Kwon, Yikun Bai, Amirhossein Mollaali, Guang Lin

Abstract: Recent data-driven methods for synthesizing 6-DoF grasp poses use generative models to learn complex grasp pose distributions and generate diverse candidate poses. In particular, SE(3)-equivariant flow-based models generate grasp poses that transform consistently with object rotations and translations. However, these methods sample by iterative numerical integration, requiring tens of function eva… ▽ More Recent data-driven methods for synthesizing 6-DoF grasp poses use generative models to learn complex grasp pose distributions and generate diverse candidate poses. In particular, SE(3)-equivariant flow-based models generate grasp poses that transform consistently with object rotations and translations. However, these methods sample by iterative numerical integration, requiring tens of function evaluations per grasp and limiting their use in real-time manipulation. We propose GraspMeanFlow, an SE(3)-equivariant MeanFlow framework for few-step 6-DoF grasp generation. Our method learns the average velocity over a finite time interval, defined through the time-ordered exponential so that it reproduces exactly the rigid-body displacement accumulated over that interval. We prove that a point-cloud-conditioned distribution transported by an equivariant average-velocity flow map remains invariant, so equivariance is retained under few-step sampling, and we condition the field on a pair of times by lifting both to equivariant vectors, leaving the backbone otherwise unchanged. For stable training, we pair a flow-matching boundary term with either of two consistency terms: the differential MeanFlow identity, whose target requires a Jacobian-vector product, or an equivalent semigroup loss that avoids it. Experiments on ACRONYM show that a single function evaluation of GraspMeanFlow reaches the EMD that an iterative SE(3) flow model needs five steps to approach, that a second instantiation of the same framework improves grasp success by up to 24.3 points in the few-step regime, and that both generate grasp distributions transforming exactly with the object. △ Less

Submitted 4 August, 2026; originally announced August 2026.

Are the Financial Reasoning from LLMs Credible? A Real World Test over Long-Horizon Statements

Authors: Xinke Tong, Xuanming Zhang, Tianyi Tang, An Yang, Jiatu Hu, Guojie Lin, Zhenzhen Shi, Lingfeng Zeng, Boyu Yang, Bing Zhao, Hu Wei, Lin Qu, Dayiheng Liu

Abstract: Do Large Language Models (LLMs) possess genuine structural reasoning, or merely rely on surface-level pattern matching? The financial domain, demanding numerical precision and multi-step logic over long contexts, is an ideal testbed. Existing benchmarks fail to capture real-world industrial complexity, predominantly relying on multiple-choice questions or single-hop QA over cropped tables while ig… ▽ More Do Large Language Models (LLMs) possess genuine structural reasoning, or merely rely on surface-level pattern matching? The financial domain, demanding numerical precision and multi-step logic over long contexts, is an ideal testbed. Existing benchmarks fail to capture real-world industrial complexity, predominantly relying on multiple-choice questions or single-hop QA over cropped tables while ignoring intricate cross-statement dynamics and temporal de-cumulation. To bridge this gap, we introduce FinIndices, a large-scale benchmark evaluating data-processing fidelity over uncropped financial statements (up to 32K tokens). Utilizing an automated synthesis pipeline with adversarial traps, FinIndices encompasses Single-Index computation and Table-Index tabulation to test complex domain, temporal, and caliber reasoning. Our evaluation reveals two severe LLM vulnerabilities. First, a "Knowledge Bottleneck": despite memorizing formulas during pre-training, models demonstrate fragile pattern matching. Removing explicit formula hints causes performance to collapse (e.g., Gemini-3.1-Pro drops from 70.70% to 38.22% on table tasks), exposing fatal flaws in temporal de-cumulation and stock-flow caliber mismatch. Second, a "Structural Bottleneck": the intense cognitive load of generating multi-metric, multi-period tables actively drains reasoning capacity. Under structural pressure, LLMs that flawlessly execute isolated derivations regress to shallow heuristics, such as fetching incorrect adjacent columns or substituting deep accounting adjustments with lazy literal arithmetic. Finally, Supervised Fine-Tuning (SFT) yields substantial zero-hint gains (+8.54% Single, +3.82% Table), validating that structured logic can be partially restored via data-centric alignment. △ Less

Submitted 22 July, 2026; originally announced July 2026.

Comments: The FinIndices dataset is publicly available at https://huggingface.co/datasets/User158072/Finindice

ACM Class: I.2.7; J.4

SE(3)-MeanFlow: Few-Step Protein Backbone Generation on Lie Groups

Authors: Yikun Bai, Binghang Lu, Yikai Liu, Elaheh Akbari, Soheil Kolouri, Linxuan Wang, Ping He, Shuchan Wang, Ruqi Zhang, Guang Lin

Abstract: Generative modeling of protein backbones promises the de novo design of proteins with prescribed structural and functional properties. Existing diffusion and flow-matching models produce high-quality backbones on SE(3)^N, but inference requires numerically integrating an ODE over hundreds of network evaluations, each involving a Lie group exponential map - a bottleneck for high-throughput design c… ▽ More Generative modeling of protein backbones promises the de novo design of proteins with prescribed structural and functional properties. Existing diffusion and flow-matching models produce high-quality backbones on SE(3)^N, but inference requires numerically integrating an ODE over hundreds of network evaluations, each involving a Lie group exponential map - a bottleneck for high-throughput design campaigns. We introduce SE(3)-MeanFlow, a few-step generative framework that extends MeanFlow from Euclidean space to the Lie group geometry of protein frames. Working natively in the Lie algebra so(3) and in R^3, we derive closed-form average-velocity identities for rotations and translations, giving simulation-free training targets. We further introduce an SE(3) alpha-Flow objective that removes the Jacobian-vector product from the rotation branch and serves as a warm-up stage, after which training switches to a small-t stabilized MeanFlow loss that is used for the remainder of pretraining and for rectification-based post-training. In protein backbone generation, SE(3)-MeanFlow matches or exceeds flow-matching baselines that use several times more sampling steps, and its advantage widens in the few-step regime, where rectification lets it lead at every matched budget - at a modest cost in diversity. △ Less

Submitted 12 August, 2026; v1 submitted 29 July, 2026; originally announced July 2026.

High-order quantum correlations in nonlinear waveguide quantum electrodynamics

Authors: I Gusti Ngurah Yudi Handayana, Ya-Tang Yu, C. -Y. Lee, K. -T. Lin, G. -D. Lin, H. H. Jen

Abstract: Quantum emitters coupled to nonlinear one-dimensional waveguides provide a route for quantum-state engineering by utilizing parametric gain accumulation along with modified waveguide-mediated interactions. Since the accumulated squeezing depends on propagation distance, different emitter separations can experience different effective gain, making the spatial structure of connected quantum correlat… ▽ More Quantum emitters coupled to nonlinear one-dimensional waveguides provide a route for quantum-state engineering by utilizing parametric gain accumulation along with modified waveguide-mediated interactions. Since the accumulated squeezing depends on propagation distance, different emitter separations can experience different effective gain, making the spatial structure of connected quantum correlations a central feature of the dynamics. Here we investigate the transient and steady-state connected correlations of emitter arrays coupled to a parametrically driven waveguide. Using an effective master equation for nonlinear waveguide QED, we analyze second- and third-order connected correlations as functions of the squeezing parameter and interparticle distance. In the two-emitter limit, accumulated squeezing drives excitation buildup and generates a nonzero connected second-order correlation. For many-emitter arrays, tuning the interparticle distance switches the dominant local second-order correlation between the bulk and boundary regions, and enables an analogous spatial control of genuine third-order connected correlations. In the steady state, the averaged second-order correlation exhibits sign-changing and nonmonotonic behavior in the squeezing--interparticle distance parameter space, whereas the third-order connected correlation is strongly enhanced by increasing the squeezing parameter. These results identify nonlinear waveguide QED as a tunable platform for spatially controlling connected quantum correlations and provide insights into correlation engineering in open quantum optical arrays. We further show that this local contrast remains visible in both odd and even number of atomic arrays. △ Less

Submitted 27 July, 2026; originally announced July 2026.

Comments: 12 pages, 6 figures

Two-Timescale Hierarchical Reinforcement Learning for Resilient Operations

Authors: Young Hyun Cho, Franz Stoll, Will Wei Sun, Guang Lin, Stephan Biller

Abstract: Unexpected shocks recur in global operations, requiring decision rules that adapt as market and operating conditions change. Many operational systems also have hierarchical structures in which long-term and short-term decisions pursue a shared objective. We study how hierarchical reinforcement learning can strengthen resilience by adapting these interdependent rules jointly. We develop a two-times… ▽ More Unexpected shocks recur in global operations, requiring decision rules that adapt as market and operating conditions change. Many operational systems also have hierarchical structures in which long-term and short-term decisions pursue a shared objective. We study how hierarchical reinforcement learning can strengthen resilience by adapting these interdependent rules jointly. We develop a two-timescale hierarchical reinforcement learning framework that adapts long-term and short-term policies at their respective time scales. Because the policies are interdependent, we synchronize their updates and prove, to our knowledge, the first convergence guarantees for coupled two-timescale learning. Over $T$ periods, our policies' average gap from an optimal policy pair is $O(T^{-1/2})$, improving to $O(\log T/T)$ when poor decisions produce clearer profit losses. In a used-car case study, inventory replenishment is the long-term decision and customer-arrival pricing the short-term decision. Relative to the strongest partially adaptive benchmark, the framework increases mean profit by $9.2\%$ under joint demand-supply shocks and by $11.8\%$ under a prolonged shock scenario, while maintaining a more stable profit trajectory over time. Short-term adaptation addresses routine seasonality and one-sided disruptions by responding immediately to changing conditions. Under joint demand-supply shocks, however, it is insufficient alone; long-term adaptation is also needed to create favorable conditions for short-term decisions. Joint adaptation thus yields higher and more stable profits through disruption and recovery. Because many organizations already use hierarchical planning, the framework strengthens operational resilience without altering existing decision structures. △ Less

Submitted 25 July, 2026; originally announced July 2026.

Hyperdeterminant wavefunctions

Authors: Guan-Lin Lin, Di Xiao, Ying Ran

Abstract: We systematically introduce hyperdeterminant wavefunctions as a variational-wavefunction-based theoretical framework for strongly correlated quantum states of matter, together with practical numerical simulation algorithms. This framework generalizes previously known fermionic parton constructions, yields reliable microscopics with intuitive physical pictures, and allows direct access to the fract… ▽ More We systematically introduce hyperdeterminant wavefunctions as a variational-wavefunction-based theoretical framework for strongly correlated quantum states of matter, together with practical numerical simulation algorithms. This framework generalizes previously known fermionic parton constructions, yields reliable microscopics with intuitive physical pictures, and allows direct access to the fractionalized degrees of freedom together with associated microscopic effective field theories. We demonstrate the applications of this framework to fractional Chern insulators and quantum spin liquids. We comment that the hyperdeterminant states belong to a more general class of variational wavefunctions: the fused Gaussian states. △ Less

Submitted 25 July, 2026; originally announced July 2026.

Comments: 81 pages, 17 figures

Geometry-aware LegONet for PDE Learning on Arbitrary Domains

Authors: Jiahao Zhang, Yueqi Wang, Guang Lin

Abstract: Learned PDE solvers often entangle governing operators with the geometry, boundary conditions, and discretization used for training. This limits reuse when the same physics is posed on new domains, and it also makes physical-law discovery geometry-dependent. We introduce Geometry-aware LegONet (gLegONet), a boundary-manifold extension of Lego-like operator learning. Physical mechanisms are pretrai… ▽ More Learned PDE solvers often entangle governing operators with the geometry, boundary conditions, and discretization used for training. This limits reuse when the same physics is posed on new domains, and it also makes physical-law discovery geometry-dependent. We introduce Geometry-aware LegONet (gLegONet), a boundary-manifold extension of Lego-like operator learning. Physical mechanisms are pretrained once as modular variational blocks on an ambient spectral domain. For a target geometry, sampled boundary constraints define an affine admissible manifold. Its mass-orthonormal tangent coordinates are used to evolve the dynamics and evaluate candidate law-discovery features directly. Changing the domain therefore changes only an algebraic coordinate interface, not the learned operator blocks. This converts arbitrary-domain PDE learning from geometry-specific retraining or soft penalty enforcement into boundary-guaranteed assembly of reusable mechanisms. In forward simulations and sparse identification tests on unseen domains, the method maintains boundary residuals near the algebraic tolerance and yields predictive governing laws from short-time observations. △ Less

Submitted 25 July, 2026; originally announced July 2026.

Some Integrability Properties of $m$-Subharmonic Functions

Authors: Genglong Lin

Abstract: Let $1\le m<n$ and let $u$ be an $m$-subharmonic function on a domain in $\mathbb{C}^n$. We study local exponential and polynomial integrability, with particular attention to the sharp polynomial exponent predicted by Błocki's conjecture. Explicit radial examples show that direct analogues of the Guan--Zhou strong openness theorem and Skoda's integrability criterion formulated in terms of the $m$-… ▽ More Let $1\le m<n$ and let $u$ be an $m$-subharmonic function on a domain in $\mathbb{C}^n$. We study local exponential and polynomial integrability, with particular attention to the sharp polynomial exponent predicted by Błocki's conjecture. Explicit radial examples show that direct analogues of the Guan--Zhou strong openness theorem and Skoda's integrability criterion formulated in terms of the $m$-Lelong number fail when $m<n$. We classify a family of radial power-logarithmic singularities and determine the exact $L^p$-integrability range for each member, including endpoint behavior. We resolve two problems posed by Benali--Ghiloufi. The normalized limit of the ball maximum always equals the $m$-Lelong number; this follows by combining their spherical-mean formula with the strong uniqueness theorem for tangents. The pointwise integrability exponent is lower semicontinuous in the base point. However, even when restricted to $SH_m$, it is not lower semicontinuous with respect to the $L^1_{\loc}$ topology. We also disprove their polynomial openness conjecture using an explicit power-logarithmic endpoint example. Finally, we introduce a scale of local Hessian-capacity conditions, denoted by $C_{m,δ}$. The volume-capacity inequality and the layer-cake formula yield $$u\in L^s_{loc}\quad\text{for every}\quad s<\frac{(m+δ)n}{n-m}.$$ The critical condition $\mathrm C_{m,0}=\mathrm C_m$ holds for negative functions of finite total Hessian mass with relatively compact deep sublevel sets, and for radial functions. More generally, functions in the energy class $\mathcal E_{p,m}$ satisfy $\mathrm C_{m,p}$, recovering the full Åhag--Czy{ż} Sobolev exponent. These results provide partial progress toward Błocki's conjecture, which has remained open for more than two decades. △ Less

Submitted 7 August, 2026; v1 submitted 23 July, 2026; originally announced July 2026.

MSC Class: 32U05; 32U25; 32W20; 35J60

Exploring the Small-scale Magnetic Fields in the Atmosphere of HD 49385 by Asteroseismic Analysis

Authors: Yuetong Wang, Yaguang Li, Yan Li, Guifang Lin, Tao Wu

Abstract: Recent asteroseismic studies have shown convincing evidences that magnetic fields may exist in the interior of some pulsating red giants. Inspired by this breakthrough, we explored the effect of small-scale magnetic fields on the p-mode oscillations in an evolved star, HD 49385. {\bf We incorporate a modified Eddington $T$-$τ$ equation that phenomenologically mimics the effect of the magnetic fiel… ▽ More Recent asteroseismic studies have shown convincing evidences that magnetic fields may exist in the interior of some pulsating red giants. Inspired by this breakthrough, we explored the effect of small-scale magnetic fields on the p-mode oscillations in an evolved star, HD 49385. {\bf We incorporate a modified Eddington $T$-$τ$ equation that phenomenologically mimics the effect of the magnetic fields in the atmosphere of HD 49385,} and calculate the frequencies of p-modes with $l=0$, 1, and 2. By comparing the calculated frequencies with the observed ones, we select two best-fit models with either GS98 or A09 chemical composition. Our best-fit models not only fit satisfactorily the observed frequencies, but also well reproduce some spectroscopically observed stellar parameters such as effective temperature and log\,$g$. Based on the two best-fit models, we have estimated that the small-scale magnetic fields possess a strength of approximately 80\,G and spread concentratively at approximately a height of 1850 km in the atmosphere. By selecting the best-fit models with special requirement on the avoided-crossing mode, we have confirmed that the frequency of the avoided-crossing mode is tightly related to the helium core of the star, and determined the size of the helium core as 0.117${\rm M}_\odot$ in mass and 0.078${\rm R}_\odot$ in radius. Based on the improvements of previous two sides, we can accurately determine the mass of HD 49385 to be $1.25\pm 0.02\,{\rm M}_\odot$ with an age of 4.1\,Gyr for GS98 composition and 4.5\,Gyr for A09 composition. △ Less

Submitted 21 July, 2026; originally announced July 2026.

Comments: has been published in APJ

Asteroseismic Analysis of a Red Giant KIC 9145955 by Including the Small-scale Magnetic Fields in the Atmosphere

Authors: Yuetong Wang, Guifang Lin, Yan Li, Tao Wu, Yaguang Li

Abstract: Recent convincing evidence is found within asteroseismology that suggests the magnetic fields exist in three red giants. Research on small-scale magnetic fields in the Sun and HD 49385 has shown that they have a certain corrective effect on the systematic discrepancies between observed and theoretical frequencies. Here we apply a similar method applied for the Sun to a red giant, KIC 9145955, to e… ▽ More Recent convincing evidence is found within asteroseismology that suggests the magnetic fields exist in three red giants. Research on small-scale magnetic fields in the Sun and HD 49385 has shown that they have a certain corrective effect on the systematic discrepancies between observed and theoretical frequencies. Here we apply a similar method applied for the Sun to a red giant, KIC 9145955, to explore the impact of small-scale magnetic fields in the photosphere on its frequencies. We find that the calculated frequencies of our best-fit model, which simulates the effect of the magnetic fields by artificially modifying the Eddington $T-τ$ relation, perfectly match those of the observed l = 0, 1, and 2 modes, indicating the existence of small-scale magnetic fields with an upper strength limit of 65 G and concentrating at a height 13,100 km in the photosphere. Based on the best-fit model, we revise the stellar parameters of KIC 9145955 as: $M = 1.23\pm0.04\,M_\odot$, $R = 5.57\pm0.06\,R_\odot$, $L = 19.85\pm0.5\,L_\odot$, $Age = 3.83\pm0.5$\,Gyr, $M_{\rm He} = 0.2108\pm0.0005M_\odot$, and $R_{\rm He} = 0.0306\pm0.0001R_\odot$. △ Less

Submitted 21 July, 2026; originally announced July 2026.

Comments: has been published in ApJ

A Bayesian Framework for Built-in Input Dimension Reduction for Gaussian Process Modeling

Authors: Eric Herrison Gyamfi, Emily L. Kang, Bledar A. Konomi, Guang Lin

Abstract: Gaussian process (GP) modeling is widely used in computational science and engineering. However, fitting a GP to high-dimensional inputs remains challenging due to the curse of dimensionality. While various methods have been proposed to reduce input dimensionality, they typically follow a two-stage approach, performing dimension reduction and GP fitting separately. We introduce a Bayesian framewor… ▽ More Gaussian process (GP) modeling is widely used in computational science and engineering. However, fitting a GP to high-dimensional inputs remains challenging due to the curse of dimensionality. While various methods have been proposed to reduce input dimensionality, they typically follow a two-stage approach, performing dimension reduction and GP fitting separately. We introduce a Bayesian framework that seamlessly integrates dimensionality reduction with GP modeling and inference. Our approach, built on a hierarchical Bayesian model with priors on the Stiefel manifold, enforces orthonormality on the projection matrix and enables posterior inference via Hamiltonian Monte Carlo with geodesic flow. Additionally, we extend this framework by incorporating Deep Gaussian Processes (DGP) with built-in dimension reduction, providing a more flexible and powerful tool for complex datasets. Through extensive numerical studies, we demonstrate that while the proposed Bayesian method incurs higher computational costs, it improves predictive performance and uncertainty quantification, providing a principled and robust alternative to existing methods. △ Less

Submitted 21 July, 2026; originally announced July 2026.

Generative World Renderer at the Speed of Play

Authors: Guixu Lin, Zheng-Hui Huang, Siqi Yang, Ming-Hsuan Yang, Kaipeng Zhang, Zhixiang Wang

Abstract: Generative world renderer AlayaRenderer receives structured world states exported from physics engines and synthesizes RGB frames. Unlike models that generate frames from text/control-hints prompts, AlayaRenderer preserves scene structure without altering the underlying world dynamics. This demonstrates an alternative path toward interactive world modeling and user-controllable play. However, the… ▽ More Generative world renderer AlayaRenderer receives structured world states exported from physics engines and synthesizes RGB frames. Unlike models that generate frames from text/control-hints prompts, AlayaRenderer preserves scene structure without altering the underlying world dynamics. This demonstrates an alternative path toward interactive world modeling and user-controllable play. However, the original AlayaRenderer is too computationally expensive for real-time deployment. This technical report introduces AlayaRenderer-Flash, a real-time-oriented generative forward world renderer that pushes AlayaRenderer from 0.56 FPS to 31.54 FPS, reaching the speed of play. AlayaRenderer-Flash reformulates the original renderer as a few-step autoregressive streaming model and introduces lightweight distilled codecs for efficient latent encoding and frame reconstruction. It retains the teacher model's G-buffer and text-prompt interfaces while enabling continuous rendering over input streams of unbounded length. We evaluate AlayaRenderer-Flash on G-buffer streams across content preservation, temporal consistency, cross-window stability, prompt controllability, and runtime efficiency. Our results show that AlayaRenderer-Flash substantially reduces inference cost while preserving the core rendering capabilities of the teacher model. By integrating AlayaRenderer-Flash with a physics engine, we build a fully playable generative world running at 30 FPS. △ Less

Submitted 21 July, 2026; originally announced July 2026.

Comments: Project page: https://alaya-renderer-flash.alayalab.ai/

Final assessment of radioactive impurities in the JUNO detector

Authors: Thomas Adam, Fengpeng An, Costas Andreopoulos, Giuseppe Andronico, Nikolay Anfimov, Vito Antonelli, Tatiana Antoshkina, João Pedro Athayde Marcondes de André, Didier Auguste, Nikita Balashov, Andrea Barresi, Davide Basilico, Eric Baussan, Marco Beretta, Antonio Bergnoli, Nikita Bessonov, Daniel Bick, Lukas Bieger, Svetlana Biktemerova, Thilo Birkenfeld, Simon Blyth, Manuel Böhles, Anastasia Bolshakova, Mathieu Bongrand, Matteo Borghesi , et al. (549 additional authors not shown)

Abstract: The Jiangmen Underground Neutrino Observatory (JUNO) collaboration has completed the construction of the 20,000-ton liquid scintillator detector and the associated muon veto detector system. To meet the physics objectives, the materials used in the detector must exhibit low radioactive contamination. The single-event rate in the fiducial volume (R $<$ 17.2 m) of the scintillator is required to be… ▽ More The Jiangmen Underground Neutrino Observatory (JUNO) collaboration has completed the construction of the 20,000-ton liquid scintillator detector and the associated muon veto detector system. To meet the physics objectives, the materials used in the detector must exhibit low radioactive contamination. The single-event rate in the fiducial volume (R $<$ 17.2 m) of the scintillator is required to be approximately 7 Hz for energies above 0.7 MeV, resulting in an accidental coincidence background of about 1 event per day for reactor neutrino physics analyses. Since the beginning of the construction phase, we have screened the natural radioactivity content of thousands of materials, to select those that meet the design background budget. The radioactive impurity concentrations of the materials ultimately used in the JUNO detector are summarized in this paper. The construction of the entire detector and the subsequent filling of the liquid scintillator were completed in August 2025. From the initial data, the total count rate of natural radioactivity within the detector's fiducial volume has met the requirements and is sufficient to support the reactor antineutrino analysis. △ Less

Submitted 19 July, 2026; originally announced July 2026.

Scaling law for optimal excitation storage and superradiant release in waveguide QED systems

Authors: Wei Chen, Kuan-Ting Lin, Guin-Dar Lin, Hsiang-Hua Jen

Abstract: Driven-dissipative quantum emitters provide a powerful platform for controllable excitation storage and release, with promising applications in quantum batteries and quantum storage. Yet, transient excitation transfer in collective many-body systems is often obscured by the intricate interplay among coherent driving, dissipation, and correlation dynamics. Here, we uncover a scalable excitation-sto… ▽ More Driven-dissipative quantum emitters provide a powerful platform for controllable excitation storage and release, with promising applications in quantum batteries and quantum storage. Yet, transient excitation transfer in collective many-body systems is often obscured by the intricate interplay among coherent driving, dissipation, and correlation dynamics. Here, we uncover a scalable excitation-storage mechanism in two emitter ensembles coupled to a semi-infinite waveguide. A coherently driven ensemble acts as an effective excitation reservoir, while a second ensemble positioned near a dissipative node serves as a subradiant storage medium. Surprisingly, when the driven ensemble largely exceeds the storage ensemble in size, the transfer dynamics enters a nearly correlation-free regime, allowing the driven ensemble to behave effectively as a classical excitation source. This reveals a simple scaling law for optimal excitation transfer, under which the storage ensemble approaches near-complete population inversion as the driven ensemble size increases. Building on this mechanism, we propose a three-stage storage-and-release protocol enabling fast excitation storage and controllable photon emission. Our results demonstrate how coherent and dissipative collective interactions can be jointly harnessed for quantum energy storage and programmable nonequilibrium dynamics in waveguide QED platforms. △ Less

Submitted 19 July, 2026; originally announced July 2026.

A Low-energy Threshold and Multi-messenger Trigger System for the JUNO Experiment

Authors: Thomas Adam, Fengpeng An, Costas Andreopoulos, Giuseppe Andronico, Nikolay Anfimov, Vito Antonelli, Tatiana Antoshkina, João Pedro Athayde Marcondes de André, Didier Auguste, Nikita Balashov, Andrea Barresi, Davide Basilico, Eric Baussan, Marco Beretta, Antonio Bergnoli, Nikita Bessonov, Daniel Bick, Lukas Bieger, Svetlana Biktemerova, Thilo Birkenfeld, Simon Blyth, Manuel Boehles, Anastasia Bolshakova, Mathieu Bongrand, Matteo Borghesi , et al. (543 additional authors not shown)

Abstract: The Jiangmen Underground Neutrino Observatory (JUNO) is a 20-kiloton liquid scintillator neutrino detector, located 650 meters (1800 m.w.e.) underground in Jiangmen, Guangdong, China. JUNO is primarily designed for reactor neutrino measurements and has been taking data since 2025. With the largest mass of its kind and an excellent energy resolution, JUNO is a leading observatory for high-precision… ▽ More The Jiangmen Underground Neutrino Observatory (JUNO) is a 20-kiloton liquid scintillator neutrino detector, located 650 meters (1800 m.w.e.) underground in Jiangmen, Guangdong, China. JUNO is primarily designed for reactor neutrino measurements and has been taking data since 2025. With the largest mass of its kind and an excellent energy resolution, JUNO is a leading observatory for high-precision measurements of MeV neutrinos. The standard global trigger system serves as the primary trigger for JUNO. We present a newly developed multi-messenger trigger system that extends the capabilities of the global trigger by providing a lower energy threshold and an independent monitoring capability. During the 2025 operation, it achieved an effective energy threshold of approximately 110 +/- 10 keV, providing a lower threshold configuration suitable for low-energy event analysis. The system shows the potential to further reduce the threshold to well below 100 keV. Based on the multi-messenger trigger system, an astrophysical monitor has been developed to receive and process external alerts from other messengers, such as gravitational-wave observations. A Transient Neutrino Burst Monitor is integrated to detect short-time-scale neutrino burst events and enables real-time monitoring of transient astrophysical phenomena. The system is sensitive to neutrino bursts from core-collapse supernovae within a distance of about 250 kpc. △ Less

Submitted 15 July, 2026; originally announced July 2026.

Comments: 29 pages, 19 figures, 3 tables

MetaView: Monocular Novel View Synthesis with Scale-Aware Implicit Geometry Priors

Authors: Yufei Cai, Xuesong Niu, Hao Lu, Kun Gai, Kai Wu, Guosheng Lin

Abstract: Current visual generation models are capable of producing high-quality content, yet they lack a coherent perception of the spatial structure. Existing generative novel view synthesis methods typically introduce explicit geometry priors, which enforce spatial consistency but inherently restrict generalization in large view changes. In contrast, recent interactive generative methods favor implicit s… ▽ More Current visual generation models are capable of producing high-quality content, yet they lack a coherent perception of the spatial structure. Existing generative novel view synthesis methods typically introduce explicit geometry priors, which enforce spatial consistency but inherently restrict generalization in large view changes. In contrast, recent interactive generative methods favor implicit scene modeling, offering greater flexibility at the cost of precise camera control and geometry consistency. In this paper, we propose MetaView, a diffusion-based monocular novel view synthesis framework that enables rendering under large view changes from a single image. Our key insight is to combine implicit geometry modeling with minimal yet essential explicit 3D cues: we incorporate implicit geometry priors from a feed-forward geometry perception network to regularize structure without imposing restrictive reconstruction pipelines, while leveraging metric depth to anchor the generation to a metric scale. This design allows MetaView to achieve both geometry consistency and precise controllability. Extensive experiments demonstrate that, under challenging monocular large viewpoint changes, MetaView significantly outperforms existing methods and exhibits superior generalization. Our code is publicly available at https://github.com/KlingAIResearch/MetaView. △ Less

Submitted 5 August, 2026; v1 submitted 13 July, 2026; originally announced July 2026.

Comments: accepted to ECCV 2026

Unified Gradient Projection: Language-Balanced Continual Learning for Multilingual Low-Resource ASR

Authors: Ziang Ren, Guodong Lin, Yuchen Ai, Kaize Tan, Wei-Qiang Zhang

Abstract: Large-scale pretrained ASR models such as Whisper exhibit strong multilingual capabilities. However, fine-tuning on low-resource languages often causes catastrophic forgetting. Although continual learning mitigates this issue, existing methods struggle to regulate cross-task interference in multilingual settings, where dominant languages bias optimization. We propose Unified Gradient Projection (U… ▽ More Large-scale pretrained ASR models such as Whisper exhibit strong multilingual capabilities. However, fine-tuning on low-resource languages often causes catastrophic forgetting. Although continual learning mitigates this issue, existing methods struggle to regulate cross-task interference in multilingual settings, where dominant languages bias optimization. We propose Unified Gradient Projection (UGP), which constrains parameter updates using reference gradients from language-balanced replay in a unified projection space. By equalizing per-language contributions in the projection, UGP reduces dominant-language bias and improves cross-lingual stability. We further show that combining gradient-level projection with data-level replay yields complementary gains in stability and plasticity. Across diverse low-resource language groups and model scales, UGP enables effective adaptation while substantially mitigating forgetting. On Whisper-large-v3, it achieves near-zero average forgetting. △ Less

Submitted 13 July, 2026; originally announced July 2026.

Comments: Accepted by Interspeech 2026

A Hyperbolic Neural Closure for M1 Radiation Transfer

Authors: Bongseok Kim, Jiahao Zhang, Johannes Krotz, Dinshaw Balsara, Ryan McClarren, Guang Lin

Abstract: In radiation transfer simulations, an M1 method achieves substantial computational savings by replacing the full angular transport equation with a low-order moment system. Because this reduced system is not closed, a closure model is required to represent the unknown higher-order moments using lower-order moments. While machine learning (ML)-based closures can improve accuracy beyond classical ana… ▽ More In radiation transfer simulations, an M1 method achieves substantial computational savings by replacing the full angular transport equation with a low-order moment system. Because this reduced system is not closed, a closure model is required to represent the unknown higher-order moments using lower-order moments. While machine learning (ML)-based closures can improve accuracy beyond classical analytic closures, unconstrained learned closures may produce non-real characteristic speeds and consequently cause numerical solver breakdown. To guarantee real eigenvalues of the Jacobian associated with ML closures, we propose a hyperbolic neural closure for the M1 radiative transfer system. Rather than directly predicting closure terms, we parameterize the Jacobian through two neural networks: (i) a symmetric matrix network and (ii) a strictly convex entropy network whose Hessian defines a positive definite symmetrizer. These components are combined to yield a Jacobian that is similar to a symmetric matrix, thereby ensuring real eigenvalues. The closure is then reconstructed by numerical integration of the learned Jacobian field along a prescribed integration path. Numerical experiments show that the proposed closure not only achieves higher closure accuracy than classical analytic closures, but also improves solution accuracy and remains stable in discontinuous Galerkin simulations for radiative transfer problems. △ Less

Submitted 11 July, 2026; originally announced July 2026.

FAST Discovery of $μ$Jy Radio Pulsations from PSR J2238+5903, Providing a DM Distance Anchor for the Candidate TeV Halo 1LHAASO J2238+5900

Authors: Jianli Zhang, Hui Zhu, Guanhong Lin, Dejia Zhou, Yuting Chu, Songzhan Chen, Min Zha, WenJun Huang, ZiWei Ou, P. H. Thomas Tam, Sha Wu, Qiang Yuan, Yi Zhang

Abstract: We report the first detection of radio pulsations from PSR J2238+5903, a gamma-ray pulsar spatially coincident with the extended TeV source 1LHAASO J2238+5900. Our 3000 s FAST L-band observation reveals a weak periodic signal at the known Fermi-LAT spin period, with $P=162.76568$ ms and $\mathrm{DM}=247.5\pm3.0~\mathrm{pc~cm^{-3}}$. The signal is independently confirmed by both FFT-based and Fast… ▽ More We report the first detection of radio pulsations from PSR J2238+5903, a gamma-ray pulsar spatially coincident with the extended TeV source 1LHAASO J2238+5900. Our 3000 s FAST L-band observation reveals a weak periodic signal at the known Fermi-LAT spin period, with $P=162.76568$ ms and $\mathrm{DM}=247.5\pm3.0~\mathrm{pc~cm^{-3}}$. The signal is independently confirmed by both FFT-based and Fast Folding Algorithm searches. The radiometer equation gives a flux density of $S_{1250}\simeq3\,μ$Jy, placing PSR J2238+5903 among the faintest radio-detected Fermi pulsars. Interpreting the DM with Galactic electron-density models gives $d_{\rm DM}=7.4\pm3.9$ kpc. At this distance, the LHAASO WCDA 39\% containment radius corresponds to a characteristic diameter of $\sim132$ pc, and the $>1$ TeV luminosity is $L_{\rm TeV}\simeq7.1\times10^{34}$ erg s$^{-1}$, about 8\% of the pulsar's spin-down power. The radio DM thus provides the first pulsar-specific distance constraint for assessing whether 1LHAASO J2238+5900 is a young relic-PWN / TeV-halo transition system. △ Less

Submitted 10 July, 2026; v1 submitted 9 July, 2026; originally announced July 2026.

Comments: 8 pages, 2 figures, comments welcome

EditVerse3D: High-Quality 3D Object Editing with Region-Aware Learning

Authors: Youtan Yin, Yanning Zhou, Jiacheng Wei, Xiaofeng Yang, Jun Zhang, Jiayang Bai, Jingwen Ye, Weidong Zhang, Guosheng Lin

Abstract: Local editing of 3D objects remains a long-standing challenge. When interacting with 3D content, humans naturally tend to specify a coarse region of interest for modification rather than defining precise editing boundaries. However, previous methods rely on fully edited 2D images, precise 3D masks, or redundant pipelines, which present a gap. To bridge this gap, we propose EditVerse3D, a novel 3D… ▽ More Local editing of 3D objects remains a long-standing challenge. When interacting with 3D content, humans naturally tend to specify a coarse region of interest for modification rather than defining precise editing boundaries. However, previous methods rely on fully edited 2D images, precise 3D masks, or redundant pipelines, which present a gap. To bridge this gap, we propose EditVerse3D, a novel 3D editing framework that enables high-quality object editing under such coarse guidance. Our approach takes as input a 3D object to be edited, a coarse 3D bounding box indicating the target region, and a reference 2D image describing the desired modification. It produces a coherent, high-fidelity edited 3D object. To facilitate this editing, we introduce a novel region-aware adaptive loss that emphasizes hard-to-learn regions and balances the objective between target and preserved areas. Complementing our loss function, we enhance model robustness and generalization through targeted data augmentations, such as training with scaled 3D masks and filtering out unrealistic editing pairs. We construct a large-scale 3D editing dataset derived from parts information. Extensive experiments demonstrate that EditVerse3D achieves superior visual quality and quantitative performance compared to existing 3D editing approaches. Please visit our project page at https://editverse3d.github.io. △ Less

Submitted 8 July, 2026; originally announced July 2026.

Comments: Accepted to ECCV 2026. Project page: https://editverse3d.github.io/

Decoupling Conversational Dynamics in Full-Duplex Spoken Models through Reinforcement Learning

Authors: Yuxin Li, Donghang Wu, Guan-Ting Lin, Hung-yi Lee, Chengwei Qin, Zhehuai Chen, Chen Chen

Abstract: Recent full-duplex spoken dialogue models have demonstrated compelling progress toward human-like interaction, enabling agents to respond with low latency, produce backchannels, and handle user barge-ins. Yet these improvements in conversational dynamics often come with weaker reasoning and instruction-following abilities, revealing a potential tension between interactive dynamics and intelligence… ▽ More Recent full-duplex spoken dialogue models have demonstrated compelling progress toward human-like interaction, enabling agents to respond with low latency, produce backchannels, and handle user barge-ins. Yet these improvements in conversational dynamics often come with weaker reasoning and instruction-following abilities, revealing a potential tension between interactive dynamics and intelligence capability. In this paper, we argue that such an intelligence--dynamics trade-off is not fundamental: conversational dynamics can instead be learned as a separate real-time decision policy from human dialogue data. To this end, we propose DuplexPO, a reinforcement learning (RL) framework that decouples when to speak from what to say. It preserves the semantic response capability of an instruction-tuned assistant, while optimizing its temporal interaction behavior over selected high-impact windows from long human conversations. To quantitatively optimize these dynamics, we formulate the Factorized Conversational Dynamics Reward (FCDR) to enable fine-grained temporal credit assignment for turn initiation, backchanneling, yielding, and regularized participation. The policy is then optimized with a GRPO-style objective. Experiments show that DuplexPO substantially improves full-duplex behaviors, including timely backchannels, smooth turn-taking, and barge-in handling, while maintaining strong reasoning and instruction-following performance. Moreover, improvements in dynamics-oriented metrics are reflected in better user experience, suggesting that optimizing conversational timing as a standalone objective can promote more natural full-duplex interaction. △ Less

Submitted 8 July, 2026; originally announced July 2026.

Synthetic-to-Real Translation for Class-Agnostic Motion Prediction

Authors: Yizheng Wu, Hongwei Fan, Kewei Wang, Ruibo Li, Xingyi Li, Xiao Song, Zhe Wang, Chenjing Ding, Dongliang Wang, Zhiguo Cao, Guosheng Lin

Abstract: Motion understanding is critical for ensuring safety and robustness in autonomous driving systems, driving increasing interest in motion prediction. A key challenge in this domain is the high cost associated with acquiring real-world motion labels. It is therefore ideal if we could transfer motion knowledge from synthetic data to real data. In this context, we explore the potential of synthetic-to… ▽ More Motion understanding is critical for ensuring safety and robustness in autonomous driving systems, driving increasing interest in motion prediction. A key challenge in this domain is the high cost associated with acquiring real-world motion labels. It is therefore ideal if we could transfer motion knowledge from synthetic data to real data. In this context, we explore the potential of synthetic-to-real translation for motion prediction (SRMP). However, the most used naive motion regression methods are notably sensitive to the synthetic-to-real domain shift, resulting in unreliable knowledge translation. To address this, we propose a novel approach integrating a motion knowledge translation framework with two key components: (1) objectness-aware motion prediction, which explicitly models the joint distribution of motion patterns and objectness priors to improve domain-invariant feature learning, and (2) objectness-aided motion enhancement, a motion label refinement mechanism that leverages learned objectness priors to filter motion noise. Furthermore, we present a physically-based pipeline for generating Motion4D, the first synthetic 4D LiDAR dataset tailored for SRMP research, addressing the lack of synthetic motion datasets. Experimental results demonstrate that our approach effectively bridges the domain gaps and yields superior performance on real scenes. △ Less

Submitted 7 July, 2026; originally announced July 2026.

SQGen: Structured Quantum Image Generation with Latent-Modulated Quantized Tensor Trains

Authors: Guang Lin, Qibin Zhao

Abstract: Generating images directly from quantum systems is an attractive but unresolved goal on NISQ hardware. Existing quantum generators face several coupled obstacles: barren plateaus that block trainability, expensive quantum circuit preparation, and hardware noise that erodes quantum information with depth. A further difficulty is producing image-scale output without a classical decoder, whose use wo… ▽ More Generating images directly from quantum systems is an attractive but unresolved goal on NISQ hardware. Existing quantum generators face several coupled obstacles: barren plateaus that block trainability, expensive quantum circuit preparation, and hardware noise that erodes quantum information with depth. A further difficulty is producing image-scale output without a classical decoder, whose use would otherwise break the end-to-end quantum advantage. We propose SQGen, a full quantum generator built on a quantized tensor train (QTT) with a latent modulation architecture. Specifically, SQGen promotes the QTT bond index of the target pixel distribution to ancilla bond qubits, so that each circuit site operates locally on a bond register plus the two physical qubits that carry the row- and column-bit of one image scale. We further introduce latent modulation: each re-uploading rotation is factorized at the angle level into a trainable main path plus an additive latent term, reducing to the trainable main path when the latent term is disabled. During training, we create a differentiable model in the classical system under gate-compatibility constraints, with a torus prior as the latent distribution. After training, every operator maps one-to-one to a native quantum gate, yielding a compact, deployable quantum circuit with no classical decoder in the inference path. Together, these design choices address the obstacles raised above. Extensive experiments on image datasets and synthetic data demonstrate that SQGen trains stably, generates images end-to-end from a shallow circuit with no classical decoder, and shows promising feasibility on real quantum hardware. △ Less

Submitted 14 July, 2026; v1 submitted 7 July, 2026; originally announced July 2026.

Escaping the Procrustean Bed: Groupwise Orthogonal Connectors for Audio-Language Models

Authors: Ho-Lam Chung, Ke-Han Lu, Yi-Cheng Lin, Guan-Ting Lin, Yiming Chen, Hung-yi Lee

Abstract: Audio-language models compress a speech encoder's output through a Querying Transformer (Q-Former) connector before feeding it to a large language model. We identify two failures in this compression. The connector's output vectors collapse to a single direction, and different speakers produce nearly indistinguishable outputs, with paralinguistic cues such as speaker identity, gender, and prosody l… ▽ More Audio-language models compress a speech encoder's output through a Querying Transformer (Q-Former) connector before feeding it to a large language model. We identify two failures in this compression. The connector's output vectors collapse to a single direction, and different speakers produce nearly indistinguishable outputs, with paralinguistic cues such as speaker identity, gender, and prosody lost along the way. Our method, ORCA, reverses this collapse by splitting the queries into groups whose outputs are constrained to point in different directions. On SAKURA multi-hop reasoning, ORCA gains 26.4 points over an identically trained 4B baseline, reaching 75.2% (vs. 49.0% for the 8B Audio Flamingo-3). At the connector level, the same change cuts query redundancy by 12x and raises cross-speaker variance by 75x. △ Less

Submitted 7 July, 2026; originally announced July 2026.

RoboDojo: A Unified Sim-and-Real Benchmark for Comprehensive Evaluation of Generalist Robot Manipulation Policies

Authors: Tianxing Chen, Yue Chen, Zixuan Li, Junyuan Tang, Kailun Su, Haoran Lu, Weijie Wan, Baijun Chen, Songling Liu, Haowen Yan, Honghao Su, Zhiyang Dou, Kaixuan Wang, Dandan Zhang, Yunze Liu, Yan Qin, Qiwei Liang, Qiwei Wu, Zijian Lin, Wenwei Lin, Yuran Wang, Minghua He, Tianshu Wu, Ruihai Wu, Jingquan Zhou , et al. (19 additional authors not shown)

Abstract: Generalist robot manipulation policies have advanced rapidly, yet existing benchmarks remain limited in systematically evaluating their capabilities. Many rely on simple, short-horizon, or skill-narrow tasks with limited capability coverage, and are often conducted only in simulation or only in the real world. Simulation enables scalable feedback but misses physical deployment challenges, while re… ▽ More Generalist robot manipulation policies have advanced rapidly, yet existing benchmarks remain limited in systematically evaluating their capabilities. Many rely on simple, short-horizon, or skill-narrow tasks with limited capability coverage, and are often conducted only in simulation or only in the real world. Simulation enables scalable feedback but misses physical deployment challenges, while real-world evaluation is costly, time-consuming, and difficult to reproduce. We introduce RoboDojo, a unified sim-and-real benchmark for comprehensive evaluation of generalist robot manipulation policies. RoboDojo includes 42 simulation tasks and 18 real-world tasks covering diverse and complementary manipulation capabilities. The simulation benchmark evaluates five dimensions: generalization, memory, precision, long-horizon execution, and open-vocabulary instruction following, while the real-world benchmark exposes policies to challenging physical-world deployment conditions. RoboDojo supports scalable evaluation through heterogeneous parallel simulation in Isaac Sim and provides RoboDojo-RealEval, a reproducible real-world evaluation system with remote cloud access, standardized hardware, scene reset, evaluation protocol, and deployment interface. Together with XPolicyLab, policies can be integrated once and evaluated across simulation and real-world settings with minimal adaptation. We integrate 30 policies into XPolicyLab and evaluate them on RoboDojo, establishing a public leaderboard and systematic analysis of current policy performance. The website is available at http://robodojo-benchmark.com/. △ Less

Submitted 8 July, 2026; v1 submitted 5 July, 2026; originally announced July 2026.

Comments: Website: https://robodojo-benchmark.com/, Code: https://github.com/RoboDojo-Benchmark/RoboDojo, Leaderboard: https://robodojo-benchmark.com/leaderboard

Constrained particle on a group: from propagators to correlators

Authors: Guanda Lin

Abstract: We develop a particle-on-a-group formulation of super-JT gravity aimed at computing supersymmetric correlators. We show that the (super)JT gravity can be described by a particle moving on the isometry group satisfying constraints from boundary conditions of (super)JT gravity. In this language the $\mathcal N=2$ and $\mathcal N=4$ theories are described by constrained particles on $SU(1,1|1)$ and… ▽ More We develop a particle-on-a-group formulation of super-JT gravity aimed at computing supersymmetric correlators. We show that the (super)JT gravity can be described by a particle moving on the isometry group satisfying constraints from boundary conditions of (super)JT gravity. In this language the $\mathcal N=2$ and $\mathcal N=4$ theories are described by constrained particles on $SU(1,1|1)$ and $PSU(1,1|2)$. Solving the constraints gives the super-Schwarzian actions. We also quantize the reduced superparticle, with careful treatment of the fermionic constraints. We then derive the physical worldline supercharges from the requirement that the transformations preserve the constraints. These charges allow us to construct supersymmetric interval propagators in invariant variables and to formulate boundary-anchored Wilson-line operators for both superconformal primary and descendant insertions. Finally, we use these ingredients to build an algorithm for correlators. We obtain the $\mathcal N=2$ and $\mathcal N=4$ three-point composition kernels and zero-energy scalar three-point functions. For four-point functions, the same method reproduces the standard bosonic JT OTOC, gives an explicit zero-energy OTOC in $\mathcal N=2$ and $\mathcal N=4$ SJT, which is potentially useful for studying Berry curvature and BPS chaos. △ Less

Submitted 29 June, 2026; originally announced June 2026.

Comments: 48pages+appendices

GigaSpeechBench: A Real-World Multilingual Speech-to-Text Benchmark

Authors: Yujie Tu, Yifan Yang, Tianrui Wang, Yanqiao Zhu, Guodong Lin, Mingchen Shao, Haoran Wang, Junzhe Liu, Yuxiang Fu, Yizhou Peng, Changsong Liu, Peng Wang, Zhikang Niu, Yunchong Xiao, Haolong Zheng, Xiuwen Zheng, Xulin Fan, Wei-Qiang Zhang, Lei Xie, Longbiao Wang, Eng-Siong Chng, Jiajun Zhang, Kele Xu, Jianwei Yu, Binbin Zhang , et al. (13 additional authors not shown)

Abstract: While modern ASR systems achieve low error rates on high-resource benchmarks, such performance often overestimates real-world robustness. Existing evaluations address challenges in isolation, lacking a unified benchmark for domain terminology, age variation, dialects, accents, and low-resource languages, particularly across the Middle East and Southeast Asia, representing over one billion under-ev… ▽ More While modern ASR systems achieve low error rates on high-resource benchmarks, such performance often overestimates real-world robustness. Existing evaluations address challenges in isolation, lacking a unified benchmark for domain terminology, age variation, dialects, accents, and low-resource languages, particularly across the Middle East and Southeast Asia, representing over one billion under-evaluated speakers. To address this gap, we introduce GigaSpeechBench, a comprehensive multilingual and multidimensional in-the-wild ASR & AST benchmark comprising 680 hours of human-annotated speech. It features five modules: (1) 12 low-resource Middle Eastern and Southeast Asian languages, plus challenging Japanese and Korean; (2) 6 Chinese dialects; (3) 6 English accents; (4) dense terminology across 12 vertical domains for Chinese and English; and (5) older adult and child speech. We further provide human-annotated Chinese and English translations for 11 languages to support AST evaluation. Extensive evaluations of leading foundation models and commercial APIs reveal significant performance degradation in these challenging settings, exposing critical evaluation blind spots. △ Less

Submitted 21 July, 2026; v1 submitted 27 June, 2026; originally announced June 2026.

A Fast-Convergence Resolution of the Stochastic Eigenproblem Using Halley's Method and the Spectral-Chaos Approach

Authors: Hugo Esquivel, Kabir Oluwatobi Idowu, Guang Lin

Abstract: Solving stochastic eigenvalue problems has long been essential for informed decision-making, advancing scientific knowledge, and ensuring the reliability of engineering designs and applications. This paper underscores the need to continue enhancing existing numerical methods for solving the stochastic eigenproblem in order to improve convergence rates, computational efficiency, and robustness. Spe… ▽ More Solving stochastic eigenvalue problems has long been essential for informed decision-making, advancing scientific knowledge, and ensuring the reliability of engineering designs and applications. This paper underscores the need to continue enhancing existing numerical methods for solving the stochastic eigenproblem in order to improve convergence rates, computational efficiency, and robustness. Specifically, we propose a novel spectral-chaos method for solving the stochastic (linear) eigenvalue problem, employing Halley's method as the root-finding algorithm to leverage its cubic convergence properties. Our method achieves maximal convergence in solving stochastic eigenvalue problems since its rate cannot be further improved using a higher-order Householder method due to the quadratic nature of the resulting system of equations. Additionally, due to the complexity of the resulting system of equations, a tensorial approach was developed to tackle the challenges associated with the dimensional multiplicity of the stochastic eigenvalue problem, without which the solution would have been intractable. The method is derived rigorously, with a detailed error analysis that highlights the benefit of using our approach when the eigenvector components are nearly known, the computational cost of the method is also rigorously presented, and an illustrative example is provided to demonstrate the implementation of the method. Subsequently, a case study is demoed to analyze the results and validate the advantages of using Halley's method over Newton's method and Monte Carlo simulations. △ Less

Submitted 24 June, 2026; originally announced June 2026.

MSC Class: 65F15; 65C20; 65H10; 65Y20

Where Larger Models Excel: The Primacy of Constraint-Guided Reasoning

Authors: Guan-Yi Lin, Hen-Hsen Huang

Abstract: Larger language models consistently outperform smaller ones on reasoning benchmarks, yet the reasoning differences underlying this gap remain underexplored. Across benchmarks in mathematics, physics, chemistry, and programming, we observe stable performance gaps: averaged over datasets, Qwen3-32B outperforms Qwen3-8B by 6.43%, while GPT-OSS-120B exceeds GPT-OSS-20B by 7.38%. To study the reasoning… ▽ More Larger language models consistently outperform smaller ones on reasoning benchmarks, yet the reasoning differences underlying this gap remain underexplored. Across benchmarks in mathematics, physics, chemistry, and programming, we observe stable performance gaps: averaged over datasets, Qwen3-32B outperforms Qwen3-8B by 6.43%, while GPT-OSS-120B exceeds GPT-OSS-20B by 7.38%. To study the reasoning differences behind these gains, we develop AdvCluster, an automated framework that identifies questions where the larger model shows a stable advantage, extracts fine-grained advantage descriptions from paired reasoning traces produced by larger and smaller models, and organizes them through semantic clustering with quantitative evaluation and selection guided by a reviewer model. Our analysis yields a systematic taxonomy of larger model reasoning advantages, spanning both common advantages that recur across domains and specialized advantages associated with particular domains. Across these patterns, a recurring theme is Constraint-Guided Reasoning: larger models are better at identifying explicit and implicit constraints, organizing them into structured reasoning, and using them to rule out infeasible paths and verify intermediate steps. △ Less

Submitted 9 May, 2026; originally announced June 2026.

Comments: 10 pages, 3 figures,

ACM Class: I.2.7

Chinese Sunspot Drawings and Their Digitization-(VIII) Release of a Machine Readable Parameters Catalog

Authors: G. H. Lin, S. Liu

Abstract: We present a systematically revised and manually verified digital catalog of historical sunspot drawing observations from Chinese observatories, encompassing sunspot number, sunspot group number, and sunspot area measurements. This represents an updated release of China's early sunspot drawings that have been utilized in numerous published studies over the years, with ongoing refinements and corre… ▽ More We present a systematically revised and manually verified digital catalog of historical sunspot drawing observations from Chinese observatories, encompassing sunspot number, sunspot group number, and sunspot area measurements. This represents an updated release of China's early sunspot drawings that have been utilized in numerous published studies over the years, with ongoing refinements and corrections progressively incorporated during the course of those studies and now consolidated through a systematic round of data revision aimed at improving internal consistency and usability for the broader research community. To facilitate scientific utilization, selected comparisons with international reference sunspot data are presented, focusing on relative sunspot numbers (RSN) and sunspot group numbers (GN) from Yunnan Observatory (YNO) and Purple Mountain Observatory (PMO), while sunspot area measurements are also included in this release but are not compared with international datasets in the present work. For RSN, both raw and k-corrected values are analyzed; for GN, only raw counts are used as no dedicated correction factor exists in the Chinese records. The results demonstrate strong positive correlations, with correlation coefficients consistently exceeding 0.95 for all comparisons, confirming the reliability and scientific value of these historical records. Systematic differences are observed: raw YNO RSN values are systematically higher than SIDC by about 17.6\%, while raw PMO RSN values are about 6.0\% lower. After k-correction, YNO shifts to a 20.9\% deficit, suggesting over-correction, while PMO remains lower with a 15.5\% deficit. Notably, GN exhibits smaller systematic offsets (16.5\% for YNO and 8.5\% for PMO) and serves as a more robust index for long-term solar cycle studies. △ Less

Submitted 4 July, 2026; v1 submitted 23 June, 2026; originally announced June 2026.

Comments: 9figs 9tables

Journal ref: ApJs 2020 06 22

Recursive behavior in a diatomic FPUT lattice

Authors: Guo Deng, Andrea Pezzi, Genghong Lin, Miguel Onorato

Abstract: We study the diatomic FPUT lattice with cubic anharmonic potential, and analyze the recurrent behaviour of its solutions. We find that two distinct types of recurrence occur. One type is the classic FPUT recurrence; for such recurrence, we find that the relation between recurrence period and nonlinear strength is similar to that in the monatomic case. The other type, which cannot exist in the mona… ▽ More We study the diatomic FPUT lattice with cubic anharmonic potential, and analyze the recurrent behaviour of its solutions. We find that two distinct types of recurrence occur. One type is the classic FPUT recurrence; for such recurrence, we find that the relation between recurrence period and nonlinear strength is similar to that in the monatomic case. The other type, which cannot exist in the monatomic lattice, is the recurrence due to the interactions between modes in the two branches of the dispersion relation. Indeed, we prove the existence of the optical-acoustical-acoustical resonant interaction between three Fourier modes for which a recurrent behavior in the distribution of the energy is observed. In addition, we develop a reduced Fourier-space dynamical model that reproduces the same recurrent behavior. We assess the robustness of our results through numerical simulations of the diatomic Toda lattice and the diatomic granular chain; in both cases, the same recursive behavior is observed. Finally, in the continuous limit, we derive from the diatomic model a system of three coupled PDEs which are known to be integrable. △ Less

Submitted 23 June, 2026; originally announced June 2026.

EvolveNav: Proactive Preflection and Self-Evolving Memory for Zero-Shot Object Goal Navigation

Authors: Qi Chai, Wenhao Shen, Nanjie Yao, Yue Xia, Kaiyong Zhao, Jie Ma, Guosheng Lin, Hao Wang

Abstract: Zero-Shot Object-Goal Navigation (ZS-OGN) requires embodied agents to explore and locate target objects without any prior training. To this end, recent methods leverage foundation models. But they typically rely on static priors and lack adaptation, which leads to repeated errors and costly trial and error. In this paper, we propose a self-evolving ZS-OGN framework that enables continuous test-tim… ▽ More Zero-Shot Object-Goal Navigation (ZS-OGN) requires embodied agents to explore and locate target objects without any prior training. To this end, recent methods leverage foundation models. But they typically rely on static priors and lack adaptation, which leads to repeated errors and costly trial and error. In this paper, we propose a self-evolving ZS-OGN framework that enables continuous test-time improvement. Specifically, we build an agentic rule memory by extracting actionable knowledge from past trajectories. Then, we propose a retrieval strategy based on upper confidence bound, selecting effective rules by balancing semantic relevance and historical success. In addition, we introduce a memory-guided preflection module that forecasts potential outcomes before action, reducing inefficient exploration. Extensive experiments show that our method outperforms existing zero-shot baselines, achieving a 10.1\% improvement in success rate with fewer unnecessary steps. △ Less

Submitted 16 June, 2026; originally announced June 2026.

Augmented Lagrangian Method for Mathematical Programs with Second-Order Cone Complementarity Constraints

Authors: Yan-Chao Liang, Chen-Yuan Zhu, Sheng-Jie Zhang, Gui-Hua Lin, Xide Zhu

Abstract: This paper investigates mathematical programs with second-order cone complementarity constraints (SOCMPCCs), which extend classical mathematical programs with complementarity constraints (MPCCs) by incorporating second-order cone structures. SOCMPCCs present significant theoretical and computational challenges, primarily due to the failure of standard constraint qualifications (such as Robinson's… ▽ More This paper investigates mathematical programs with second-order cone complementarity constraints (SOCMPCCs), which extend classical mathematical programs with complementarity constraints (MPCCs) by incorporating second-order cone structures. SOCMPCCs present significant theoretical and computational challenges, primarily due to the failure of standard constraint qualifications (such as Robinson's constraint qualification) at all feasible points. This difficulty hinders the direct application of classical nonlinear programming theories and algorithms. Motivated by the success of the augmented Lagrangian method (ALM) in solving MPCCs, we explore its extension to SOCMPCCs. The ALM, known for its matrix-free implementation and strong local convergence properties, is well suited for handling the intricate interplay between complementarity and second-order cone constraints. In this paper, we propose a tailored ALM algorithm framework for SOCMPCCs and establish its feasibility and convergence properties. We show that, under bounded ALM penalty parameters or bounded augmented Lagrangian functions, the generated sequence converges to feasible points of the SOCMPCC. Furthermore, under feasibility and additional SOCMPCC-nondegeneracy condition, we prove convergence to K-stationary points, which constitute a fundamental optimality condition for SOCMPCCs. Numerical experiments, including both illustrative examples and high-dimensional problems, are conducted to demonstrate the effectiveness and practical applicability of the proposed algorithm in addressing the challenges inherent in SOCMPCCs. △ Less

Submitted 15 June, 2026; originally announced June 2026.

Exploring How Agent Voice Accents Shape Human-AI Collaboration in K-12 Group Learning

Authors: Prerna Ravi, Carúmey Stevens, Ben Hurt, Brandon Hanks, Grace Lin, Emma Anderson

Abstract: Collaboration is widely recognized as a cornerstone of 21st-century education, yet teachers still encounter persistent challenges in fostering productive peer interaction. LLM conversational peer agents introduce new possibilities for mediating in-person group work, raising questions about how persona design, particularly their voice characteristics, shapes learners' perceptions, trust, and intera… ▽ More Collaboration is widely recognized as a cornerstone of 21st-century education, yet teachers still encounter persistent challenges in fostering productive peer interaction. LLM conversational peer agents introduce new possibilities for mediating in-person group work, raising questions about how persona design, particularly their voice characteristics, shapes learners' perceptions, trust, and interactional dynamics. While prior work has examined agent accent effects in one-to-one settings, little is known about how these effects manifest in groups. We conducted a between-subjects mixed-methods study with 33 teachers examining how a GenAI voice agent with different accents (British, Indian, and African American) influenced collaboration and agent perception. Across surveys, group interaction analyses, and artifacts, we find that accent shaped participants' mental models and the roles the agent assumed in group interaction. The British-accented agent was largely treated as a tool and engaged in detached, utility-based ways, whereas Indian- and African American-accented agents were more readily anthropomorphized and integrated as peers. These role expectations influenced trust, engagement, and reliance over time. This work advances understanding of how GenAI's sociolinguistic design features shape group dynamics in CSCL, with implications for designing culturally inclusive AI partners in group learning. △ Less

Submitted 10 June, 2026; originally announced June 2026.

Building Social World Models with Large Language Models

Authors: Haofei Yu, Yining Zhao, Guanyu Lin, Jiaxuan You

Abstract: Understanding and predicting how social beliefs evolve in response to events -- from policy changes to scientific breakthroughs -- remains a fundamental challenge in social science. Given LLMs' commonsense knowledge and social intelligence, we ask: Can LLMs model the dynamics of social beliefs following social events? In this work, we introduce the concept of the Social World Model (SWM), a genera… ▽ More Understanding and predicting how social beliefs evolve in response to events -- from policy changes to scientific breakthroughs -- remains a fundamental challenge in social science. Given LLMs' commonsense knowledge and social intelligence, we ask: Can LLMs model the dynamics of social beliefs following social events? In this work, we introduce the concept of the Social World Model (SWM), a general framework designed to capture how social beliefs evolve in response to major events. SWM learns state-transition functions for social beliefs by mining temporal patterns in social data and optimizing the evidence lower bound, without the need for explicit human annotations linking events to belief shifts, or for expensive census data. To evaluate SWM, we introduce a benchmark, SWM-bench, derived from real-world prediction markets, specifically Kalshi and Polymarket. SWM-bench includes over 12k data points for social belief prediction tasks spanning diverse domains such as politics, finance, and cryptocurrency. Our experimental results show that SWM significantly outperforms time-series foundation models, achieving state-of-the-art results on Kalshi data and demonstrating competitive performance on Polymarket data, while offering interpretable insights into the underlying mechanisms of social belief dynamics. △ Less

Submitted 9 June, 2026; originally announced June 2026.

Comments: 9 pages. ICML 2026

Enhancing Multilingual LLM-based ASR with Mixture of Experts and Dynamic Downsampling

Authors: Guodong Lin, Ziqi Chen, Yuxiang Fu, Ke Li, Wei-Qiang Zhang

Abstract: The rapid progress of large language models (LLMs) has opened up a new frontier for automatic speech recognition (ASR), making their effective integration a critical and challenging research direction. To this end, this work proposes a projector-based LLM-ASR framework targeting the key challenges of multilingual generalization and modality alignment. Our approach incorporates a Mixture of Experts… ▽ More The rapid progress of large language models (LLMs) has opened up a new frontier for automatic speech recognition (ASR), making their effective integration a critical and challenging research direction. To this end, this work proposes a projector-based LLM-ASR framework targeting the key challenges of multilingual generalization and modality alignment. Our approach incorporates a Mixture of Experts (MoE) architecture to improve cross-lingual adaptability, and a Continuous Integrate-and-Fire (CIF) mechanism for dynamic downsampling and modality alignment. Experimental results show that the combination of these components yields substantial performance improvements, surpassing strong baseline models. The proposed method represents a step toward building more accurate, robust, and generalizable LLM-based ASR systems. △ Less

Submitted 9 June, 2026; originally announced June 2026.

Comments: Accepted by ICASSP 2026

Journal ref: ICASSP (2026),18807-18811

Fine-Tuning Diffusion Models for Molecular Generation via Reinforcement Learning and Fast Sampling

Authors: Guang Lin, Shikui Tu, Lei Xu

Abstract: Generating molecules that simultaneously satisfy drug-like properties and conform to the 3D structure of a target protein is a core challenge in structure-based drug design (SBDD). Existing generative approaches, however, often rely on costly post-hoc processing during Sampling or require carefully curated datasets during training, yet still achieve modest gains. These limitations are especially p… ▽ More Generating molecules that simultaneously satisfy drug-like properties and conform to the 3D structure of a target protein is a core challenge in structure-based drug design (SBDD). Existing generative approaches, however, often rely on costly post-hoc processing during Sampling or require carefully curated datasets during training, yet still achieve modest gains. These limitations are especially pronounced in multi-objective settings, where balancing conflicting criteria remains a core challenge. To address these challenges, We propose FTDiff, a reinforcement learning fine-tuning framework tailored for diffusion-based molecular generation under structural constraints. To ensure stable and sample-efficient optimization, FTDiff adopts a group relative policy optimization (GRPO) style strategy. Furthermore, FTDiff builds upon a time-free pretrained diffusion model and incorporates a fast sampling mechanism that reduces the number of denoising steps, significantly accelerating both training and inference while maintaining generation quality. By optimizing a fixed threshold-aware reward, FTDiff effectively guides the model to produce valid, diverse, and high- quality molecules that balance multiple drug design objectives. Extensive experiments on benchmark datasets demonstrate that FTDiff consistently outperforms prior methods, without requiring expensive post-hoc optimization or intricate data engineering. △ Less

Submitted 31 May, 2026; originally announced June 2026.

Comments: 13 pages, 7 figures

Spectral Anatomy of Quantum Gaussian Process Kernels

Authors: Jian Xu, Chao Li, Guang Lin, Yuning Qiu, Delu Zeng, John Paisley, Qibin Zhao

Abstract: Two recent results have reshaped quantum Gaussian processes (QGPs). On the one hand, \citet{lowe2025assessing} rule out the exponential speedups claimed by HHL-based QGP regression in the typical, well-conditioned regime; on the other, an independent line of work shows that highly expressive quantum kernels suffer posterior pathologies that break Bayesian optimization. We show that these seemingly… ▽ More Two recent results have reshaped quantum Gaussian processes (QGPs). On the one hand, \citet{lowe2025assessing} rule out the exponential speedups claimed by HHL-based QGP regression in the typical, well-conditioned regime; on the other, an independent line of work shows that highly expressive quantum kernels suffer posterior pathologies that break Bayesian optimization. We show that these seemingly unrelated phenomena are governed by the same quantity: the normalized spectral entropy $S(K)/\log n$ of the kernel Gram matrix. We prove a Cauchy--Schwarz tail bound on Nyström approximation error, a finite-sample variance-contraction identity in terms of Bach's degrees of freedom $d_σ(K)$, and a characterization of the \emph{target-dependent} optimal entropy via the intrinsic dimension of the target in the kernel eigenbasis. Empirically, the diagnostic is kernel-agnostic: hardware-efficient, matchgate, IQP \emph{and} RBF/Matérn/RFF/deep-kernel families all collapse onto identical $S/\log n$ curves on dequantization, ECE, and variance-contraction panels. The NLL sweet spot lives at high entropy for smooth targets and at low entropy for band-limited quantum-data targets. The diagnostic transfers from simulator to IBM Heron hardware with median absolute error $3.2\%$ and mean $5.2\%$ in $S/\log n$ across $24$ configurations at $n_q = 4$, with matchgate and IQP within $5\%$ mean and a single HE configuration returning a $30\%$ outlier that drops to $0.5\%$ on rerun (attributed to calibration drift); the same diagnostic transfers to a second Heron backend (mean error $2.7\%$) and to a $n_q = 6$ scale-up on the original backend (mean error $1.7\%$). No error mitigation is applied throughout. △ Less

Submitted 2 June, 2026; v1 submitted 29 May, 2026; originally announced May 2026.

Quantum State of a Gravitating Region

Authors: Raphael Bousso, Sami Kaya, Guanda Lin, Arvin Shahbazi-Moghaddam

Abstract: We propose that any compact $d$-manifold with elliptic data, $\mathcal{J}$, prepares a quantum state $|\mathcal{J}\rangle$ on its $(d-1)$-boundary $σ$. Elliptic data consists of metric and field values, or their conjugates, but not both. No asymptotic structure is required. Inner products and traces are evaluated by the gravitational path integral with closed boundary conditions obtained by gluing… ▽ More We propose that any compact $d$-manifold with elliptic data, $\mathcal{J}$, prepares a quantum state $|\mathcal{J}\rangle$ on its $(d-1)$-boundary $σ$. Elliptic data consists of metric and field values, or their conjugates, but not both. No asymptotic structure is required. Inner products and traces are evaluated by the gravitational path integral with closed boundary conditions obtained by gluing elliptic data manifolds. In particular, we give a prescription for the Rényi entropies $S_n$ of a subregion of $σ$. In a class of examples, we find that $S_n$ is nonnegative and nonincreasing with $n$, as required for consistency. We obtain the von Neumann entropy by analytic continuation and find agreement with the minimal surface prescription of Bousso and Penington. △ Less

Submitted 27 May, 2026; originally announced May 2026.

Comments: 6 pages, 2 figures

Covering vertices by sequential stars

Authors: Mengyuan Hu, An Zhang, Yong Chen, Zhikai Chen, Wei Ding, Guohui Lin, Jiaxuan Ma, Yue Sun

Abstract: We study the problem of covering the maximum number of vertices in a graph by a collection of vertex-disjoint stars, each with a number of satellites in a given interval $[k, \ell]$, where $1 \le k < \ell$ and $\ell$ can be infinity. This is referred to as sequential {\sc $[k, \ell]$-Star Packing} problem. It is solvable in polynomial time when $k = 1$, but becomes strongly NP-hard when $k \ge 2$.… ▽ More We study the problem of covering the maximum number of vertices in a graph by a collection of vertex-disjoint stars, each with a number of satellites in a given interval $[k, \ell]$, where $1 \le k < \ell$ and $\ell$ can be infinity. This is referred to as sequential {\sc $[k, \ell]$-Star Packing} problem. It is solvable in polynomial time when $k = 1$, but becomes strongly NP-hard when $k \ge 2$. In this paper, we propose either the first or an improved approximation algorithm for the following four sequential settings: 1) a $\frac {k+1}2$-approximation algorithm when $k \ge 3$ and $\ell = \infty$, improving the previous best ratio of $\frac {(k+1)^2}{2k+1}$; 2) a $\frac 43$-approximation algorithm when $k = 2$ and $\ell = \infty$, improving the previous best ratio of $\frac 32$; 3) the first $(1 + \frac \ell{\ell+1})$-approximation algorithm when $2 = k < \ell$; and 4) the first $(1 + \max\left\{\frac {k-1}2, \frac {(k+1) \ell}{3 (\ell+1)}\right\})$-approximation algorithm when $3 \le k < \ell$. Besides the main algorithmic techniques being local search coupled with amortized analysis, we observe augmenting configurations to bridge two distant neighborhoods for a local improvement operation. Additionally, the problem has been shown APX-hard when $k \ge 3$; we prove its APX-hardness for the last remaining case where $k = 2$. △ Less

Submitted 23 May, 2026; originally announced May 2026.

Comments: 24 pages; submitted for publication

Diff-Instruct with Diffused Reward: Towards Principled One-step Generator RL

Authors: Junyi Wu, Weijian Luo, Haoyang Zheng, Ruizhe Zhang, Guang Lin

Abstract: Recent advances in one-step text-to-image generation have enabled real-time synthesis with remarkable efficiency and quality. Previous reinforcement learning methods for one-step generators combine image-space reward optimization with diffusion noisy-space distribution matching. This paradigm brings challenges due to a mismatch between terminal reward optimization and the underlying generative dyn… ▽ More Recent advances in one-step text-to-image generation have enabled real-time synthesis with remarkable efficiency and quality. Previous reinforcement learning methods for one-step generators combine image-space reward optimization with diffusion noisy-space distribution matching. This paradigm brings challenges due to a mismatch between terminal reward optimization and the underlying generative dynamics. As a result, optimization tends to exploit stochastic degrees of freedom, often improving reward at the expense of image fidelity. To address this issue, we propose Diff-Instruct with Diffused Reward (DIDR), a data-free trajectory-level alignment framework derived from Integral KL minimization. DIDR propagates the RLHF-optimal reward-tilted clean-image distribution across all noise levels along the diffusion trajectory. We show that this objective admits the same minimizer as clean-image RLHF, while naturally inducing the Diffused Reward Score (DRS), which acts as a reward-driven correction to the reference score function. To make this practical, we further introduce the Diffused Reward Proxy (DRP), an efficient estimator of DRS based on differentiable short-step denoising. Extensive experiments demonstrate that DIDR consistently Pareto-dominates existing one-step SDXL baselines. Moreover, when transferred to a 6B DiT backbone (Z-Image), DIDR surpasses its 50-step teacher in preference alignment while requiring only a single generation step. △ Less

Submitted 26 May, 2026; v1 submitted 18 May, 2026; originally announced May 2026.

Comments: author list correction

DrawVideo: Generating Long Video from Storyboard Keyframe Sketches

Authors: Chuanzhi Xu, Huiqi Liang, Bang Shi, Huiming Zhang, Yifan Xiao, Guangcheng Lin, Haodong Chen, Qiang Qu, Zhicheng Lu, Weidong Cai

Abstract: Long video generation requires high-fidelity synthesis, coherent narrative structure, and user control over extended time spans. Existing text-to-video methods often rely on a single long prompt, limiting control over pose, composition, layout, and motion. We propose DrawVideo, a sketch-guided, storyboard-driven framework for controllable long-video generation. DrawVideo decomposes long videos int… ▽ More Long video generation requires high-fidelity synthesis, coherent narrative structure, and user control over extended time spans. Existing text-to-video methods often rely on a single long prompt, limiting control over pose, composition, layout, and motion. We propose DrawVideo, a sketch-guided, storyboard-driven framework for controllable long-video generation. DrawVideo decomposes long videos into independently controllable shots, each defined by a black-and-white sketch, an appearance prompt, and a motion prompt. The sketch controls pose and layout, the appearance prompt defines identity, scene, and style, and the motion prompt guides temporal dynamics. DrawVideo follows a hierarchical 'global multi-shot, local single-sketch' strategy: it first generates a structure-aligned reference keyframe, then expands the motion prompt into derivative keyframes representing action states, and finally synthesizes clips between adjacent keyframes to build each shot. We also introduce SketchLongVideo, the first dataset for sketch-guided text-to-long-video generation, constructed from animation videos via shot detection, keyframe extraction, vision-language recognition, prompt decomposition, and sketch conversion. Experiments show that DrawVideo achieves strong structural controllability, appearance consistency, visual stability, and coherent long-video generation. △ Less

Submitted 22 May, 2026; originally announced May 2026.

Comments: 45 pages, 19 figures