https://mghnasiri.github.io/PORID/ Latest publications in Operations Research — aggregated from arXiv, Crossref, OpenAlex, and Semantic Scholar. en-us Wed, 08 Apr 2026 08:17:28 +0000 PORID Pipeline https://doi.org/10.1007/s11590-026-02292-y https://doi.org/10.1007/s11590-026-02292-y Malaguti, E., Monaci, M., Pruente, J. Abstract We study stochastic optimization problems where both the objective function and the feasible set are subject to uncertainty. To address these challenges, we adopt a K -adaptability approach, in which K candidate solutions are computed prior to the realization of uncertainty, and the best among them is selected once the actual scenario is known. We introduce a simple low Tue, 07 Apr 2026 00:00:00 +0000 stochastic Optimization Letters https://doi.org/10.1007/s11590-026-02293-x https://doi.org/10.1007/s11590-026-02293-x Majewski, M. Tue, 07 Apr 2026 00:00:00 +0000 general-or Optimization Letters https://doi.org/10.1287/opre.2023.0167 https://doi.org/10.1287/opre.2023.0167 Liang, Y., Mao, X., Wang, S. Learning to Optimize Assortment and Inventory Decisions with Unknown Demand How should a retailer coordinate product assortment and inventory when customer preferences need to be learned on the fly and stockouts dynamically reshape demand? In the paper “Online Joint Assortment-Inventory Optimization under MNL Choices,” the authors study an online learning setting where customers follow a multinomial logit choice model with unknown parameters. The retailer repeatedly selects ass Tue, 07 Apr 2026 00:00:00 +0000 supply-chain Operations Research https://doi.org/10.1287/opre.2025.1704 https://doi.org/10.1287/opre.2025.1704 Wang, T., Xiao, L., Xu, F. Preserving Multimodular Structure in Substitution Models Substitution is common in operations problems, where one product or resource may be used to satisfy demand intended for another. In models with multiple resources and demand classes, this flexibility makes it harder to establish structural properties of optimal dynamic decisions. In “Preservation of Multimodularity: New Results and Applications,” Tong Wang, Li Xiao, and Fen Xu develop two new preservation results for mult Tue, 07 Apr 2026 00:00:00 +0000 supply-chain Operations Research https://doi.org/10.1287/opre.2025.2228 https://doi.org/10.1287/opre.2025.2228 Zhou, Z., Wang, T., Zhang, J. A New Structural Insight into Stockout-Based Substitution Stockout-based substitution creates complex stochastic dynamics in inventory systems even in highly symmetric settings. In this paper, Zhou, Wang, and Zhang examine a canonical setting in which a firm allocates a fixed inventory across multiple perfectly substitutable product types (e.g., colors or designs) and customers purchase uniformly from available options. Despite the model’s symmetry, the induced stochastic dynam Tue, 07 Apr 2026 00:00:00 +0000 supply-chain Operations Research https://doi.org/10.1287/mnsc.2025.00682 https://doi.org/10.1287/mnsc.2025.00682 Hagen, J., Hodor, M., Hurwitz, A. This study examines how a first-time malignant cancer diagnosis, acting as an informational shock to perceived longevity, affects the demand for life annuities. Using a quasi-experimental design and exploiting Swedish administrative data, we show that receiving a cancer diagnosis close to retirement reduces annuitization rates by 5.5%. The diagnosis lowers the money’s worth ratio of a life annuity by 33%, representing a substantial financial loss. Combined with the modest behavioral response, th Tue, 07 Apr 2026 00:00:00 +0000 general-or Management Science https://doi.org/10.1287/mnsc.2025.01549 https://doi.org/10.1287/mnsc.2025.01549 Cui, Y., Hong, Y., Huang, N., Wang, Y. Economists and econometricians typically use aggregate economic and financial variables for gross domestic product (GDP) prediction. However, aggregation often results in a loss of valuable information, diminishing key features such as heterogeneity, interactions, nonlinearity, and structural breaks. We propose a novel microforecasting approach, using large panel data of firm accounting earnings from corporate financial reports to forecast GDP. By employing machine learning methods, we can effec Tue, 07 Apr 2026 00:00:00 +0000 ml-for-or Management Science https://doi.org/10.1287/mnsc.2024.07656 https://doi.org/10.1287/mnsc.2024.07656 Somogyi, F. Over 85% of all foreign exchange (FX) transactions involve the U.S. dollar, whereas the United States accounts for a much smaller fraction of global economic activity. My paper attributes the dominance of the dollar in FX trading to strategic avoidance of price impact. Using a novel identification strategy, I show that, on average, 13% of the volume in dollar pairs arises from using the dollar as a vehicle currency to indirectly exchange two nondollar currencies. To rationalise this result, I de Tue, 07 Apr 2026 00:00:00 +0000 general-or Management Science https://doi.org/10.1007/s10479-026-07169-y https://doi.org/10.1007/s10479-026-07169-y Chen, W. Tue, 07 Apr 2026 00:00:00 +0000 supply-chain Annals of Operations Research https://doi.org/10.1007/s10479-026-07188-9 https://doi.org/10.1007/s10479-026-07188-9 Long, F., Tang, W., Wang, Y., Zhang, J. Tue, 07 Apr 2026 00:00:00 +0000 general-or Annals of Operations Research https://doi.org/10.1007/s10479-026-07081-5 https://doi.org/10.1007/s10479-026-07081-5 Bonilla, C., Roche, H., Vergara, M. Tue, 07 Apr 2026 00:00:00 +0000 general-or Annals of Operations Research https://doi.org/10.1007/s10479-026-07166-1 https://doi.org/10.1007/s10479-026-07166-1 Lamantia, F., Radi, D., Sushko, I. Abstract We examine environmental protection in an oligopolistic setting in which a Pigouvian pollution tax is levied whenever environmental degradation exceeds a certain threshold set by a regulator. This threshold mechanism defines two scenarios, one with low environmental degradation in which firms do not internalize pollution, and one with high degradation in which firms are involved not only in production activities but also in abatement to ease the bu Tue, 07 Apr 2026 00:00:00 +0000 stochastic Annals of Operations Research https://doi.org/10.1007/s10107-026-02352-2 https://doi.org/10.1007/s10107-026-02352-2 Lan, G., Ouyang, Y., Zhang, Z. Abstract We propose novel optimal and parameter-free algorithms for computing an approximate solution with small (projected) gradient norm. Specifically, for computing an approximate solution such that the norm of its (projected) gradient does not exceed $$\varepsilon $$ ε Tue, 07 Apr 2026 00:00:00 +0000 general-or Mathematical Programming https://doi.org/10.1007/s10107-026-02356-y https://doi.org/10.1007/s10107-026-02356-y Ju, C., Lan, G. Abstract This paper proposes a novel termination criterion, termed the advantage gap function, for finite state and action Markov decision processes (MDP) and reinforcement learning (RL). By incorporating this advantage gap function into the design of step size rules and deriving a new linear rate of convergence that is independent of the stationary state distribution of the optimal policy, we demonstrate that policy gradient methods can solve MDPs in strongly-polynomial time. Tue, 07 Apr 2026 00:00:00 +0000 ml-for-or Mathematical Programming https://doi.org/10.1007/s10107-026-02354-0 https://doi.org/10.1007/s10107-026-02354-0 O’Leary-Roseberry, T., Bollapragada, R. Abstract We consider minimizing finite-sum objective functions via Hessian-averaging based subsampled Newton methods. These methods allow for gradient inexactness and have fixed per-iteration Hessian approximation costs. The recent work (Na et al. 2023) demonstrated that Hessian averaging can be utilized to achieve fast $$\mathcal {O}\left( \sqrt{\tfrac{\log k}{k}}\right) $$ Tue, 07 Apr 2026 00:00:00 +0000 general-or Mathematical Programming https://doi.org/10.1002/nav.70069 https://doi.org/10.1002/nav.70069 Zhu, X., Sun, X. ABSTRACT This paper develops an optimal incentive compensation scheme for a project with a predetermined target but no fixed deadline. A principal sponsors the project and hires an agent to execute it, offering a lump‐sum payment that depends only on the project's completion time. The agent exerts a baseline effort level but may increase effort at a personal cost to accelerate progress, balancing the reward from completion against the cost of additional effort. The principal ai Tue, 07 Apr 2026 00:00:00 +0000 general-or Naval Research Logistics http://arxiv.org/abs/2604.06158v1 http://arxiv.org/abs/2604.06158v1 Lukas-Benedikt Fiechtner, Jose Blanchet We study, to our knowledge, the first tractable multistage ex-ante distributionally robust regret optimization (DRRO) formulation for stochastic control. We consider finite-horizon LQR under common stage-law ambiguity: disturbances are independent across time but share an unknown stage law whose mean and covariance lie in a Gelbrich ball around nominal parameters. Unlike the single-stage quadratic case, the nominal certainty-equivalent (CE) controller is generally not regret-optimal, because reu Tue, 07 Apr 2026 00:00:00 +0000 math.OC stochastic arXiv http://arxiv.org/abs/2604.06089v1 http://arxiv.org/abs/2604.06089v1 Yunxiao Ren, Dingguo Liang, Yuezu Lv, Zhisheng Duan This paper investigates the leader-following consensus problem for a class of multi-agent systems subject to adversarial attack-like external inputs. To address this, we formulate the robust leader-following control problem as a global coalitional min-max zero-sum game using differential game theory. Specifically, the agents' control inputs form a coalition to minimize a global cost function, while the attacks form an opposing coalition to maximize it. Notably, when these external adversarial at Tue, 07 Apr 2026 00:00:00 +0000 eess.SY game-theory arXiv http://arxiv.org/abs/2604.06078v1 http://arxiv.org/abs/2604.06078v1 Michele Mascherpa, Victor Molnö, Carsten Skovmose Kallesøe, Johan Karlsson In this work, we study a discrete Schrödinger bridge problem with partial marginal observations. A main difficulty compared to the classical Schrödinger bridge formulation is that our problem is not strictly convex and standard Sinkhorn-type methods cannot be directly applied. To address this issue, we propose a scalable computational method based on an entropic proximal scheme. Furthermore, we develop a framework for this problem that includes duality results, characterization of the optimal so Tue, 07 Apr 2026 00:00:00 +0000 math.OC general-or arXiv http://arxiv.org/abs/2604.06045v1 http://arxiv.org/abs/2604.06045v1 Tren Baltussen, Nathan P. Lawrence, Alexander Katriniok, Ali Mesbah, Maurice Heemels Dual control addresses the trade-off between exploitation and exploration, where control inputs both regulate the system and generate informative data for estimation and identification. For certain problem classes, control and estimation can be designed independently without loss of optimality, a property known as the separation principle. However, in stochastic control problems with model uncertainty and constraints, this principle generally breaks down, and introduces the need for dual control Tue, 07 Apr 2026 00:00:00 +0000 math.OC general-or arXiv http://arxiv.org/abs/2604.06039v1 http://arxiv.org/abs/2604.06039v1 Zhichao Jia, Guanghui Lan Value iteration-type methods have been extensively studied for computing a nearly optimal value function in reinforcement learning (RL). Under a generative sampling model, these methods can achieve sharper sample complexity than policy optimization approaches, particularly in their dependence on the discount factor. In practice, they are often employed for offline training or in simulated environments. In this paper, we consider discounted Markov decision processes with state space S, action spa Tue, 07 Apr 2026 00:00:00 +0000 math.OC ml-for-or arXiv http://arxiv.org/abs/2604.05977v1 http://arxiv.org/abs/2604.05977v1 Georgios Vasileiou, Lantian Zhang, Silun Zhang Incentive design constitutes a foundational paradigm for influencing the behavior of strategic agents, wherein a system planner (principal) publicly commits to an incentive mechanism designed to align individual objectives with collective social welfare. This paper introduces the Regret-Minimizing Adaptive Incentive Design (RAID) problem, which aims to synthesize incentive laws under information asymmetry and achieve asymptotically minimal regret compared to an oracle with full information. To t Tue, 07 Apr 2026 00:00:00 +0000 math.OC game-theory arXiv http://arxiv.org/abs/2604.05967v1 http://arxiv.org/abs/2604.05967v1 Noa Kaplan, Alberto Padoan, Anastasia Bizyaeva Understanding how training shapes the geometry of recurrent network dynamics is a central problem in time-series modeling. We study the emergence of low-dimensional dominant manifolds in the training of Reservoir Computing (RC) networks for temporal forecasting tasks. For a simplified linear and continuous-time reservoir model, we link the dimensionality and structure of the dominant modes directly to the intrinsic dimensionality and information content of the training data. In particular, for t Tue, 07 Apr 2026 00:00:00 +0000 cs.LG decomposition arXiv http://arxiv.org/abs/2604.05913v1 http://arxiv.org/abs/2604.05913v1 Joonas Lahtinen, Alexandra Koulouri Electroencephalography (EEG) source imaging aims to reconstruct the spatial distribution of neural activity within the brain from non-invasive scalp measurements. This inverse problem is severely ill-posed due to the low spatial resolution of EEG and the presence of measurement noise, necessitating robust regularization techniques. Bayesian approaches provide a principled framework for incorporating prior knowledge into the solution, where regularization naturally arises through prior distributi Tue, 07 Apr 2026 00:00:00 +0000 math.NA general-or arXiv http://arxiv.org/abs/2604.05911v1 http://arxiv.org/abs/2604.05911v1 Yuxuan Chen, Shengquan Xiang, Zhifei Zhang We prove the exponential convergence to a unique invariant measure for locally damped nonlinear Schrödinger equations, perturbed by bounded noise acting on only two Fourier modes. To tackle the lack of smoothing effect, we introduce asymptotic compactness of linearized system to enhance the coupling method. Inspired by [14,33,39], we establish a new criterion for exponential mixing. Elements from global stability, nonlinear smoothing, and geometric control are combined when applying this criteri Tue, 07 Apr 2026 00:00:00 +0000 math.AP general-or arXiv http://arxiv.org/abs/2604.05882v1 http://arxiv.org/abs/2604.05882v1 Louis Shuo Wang In this note, we develop the first-order theory of optimal control problems with box constraints on the control. We emphasize the precise modification of Pontryagin's maximum principle when the admissible control set is compact, the projection/clamping formula for scalar quadratic Hamiltonians, the distinction between intrinsic projection inside the optimality system and post hoc truncation of an unconstrained solution, and the corresponding forward-backward sweep implementation. The presentatio Tue, 07 Apr 2026 00:00:00 +0000 math.OC general-or arXiv http://arxiv.org/abs/2604.05811v1 http://arxiv.org/abs/2604.05811v1 Dongzhe Zheng, Wenjie Mei Direct collocation for Bolza optimal control yields discrete Karush-Kuhn-Tucker (KKT) points, while practical solvers expose only discrete quantities such as primal-dual iterates, reduced Hessians, and Jacobians. This creates a gap between continuous second-order optimality theory and what can be certified from solver output. We develop an a posteriori certification framework that bridges this gap. Starting from a discrete KKT solution, we reconstruct piecewise polynomial state, control, and cos Tue, 07 Apr 2026 00:00:00 +0000 math.OC decomposition arXiv http://arxiv.org/abs/2604.05685v1 http://arxiv.org/abs/2604.05685v1 Yaxin Feng, Yang Xiang, Haomin Zhou In this paper, we propose an initial value fomulation of the discrete mean field games on finite graphs (Graph MFG), and design a neural network based approach to solve it. Graph MFG describes infinite, non-cooperative and interactive homogeneous agents move on node states through the edges to optimize their own goals. Nash Equilibrium of the Graph MFG is characterized by a coupled ordinary differential equations (ODE) system, including the discrete forward continuity equation and the discrete b Tue, 07 Apr 2026 00:00:00 +0000 math.NA ml-for-or game-theory arXiv http://arxiv.org/abs/2604.05678v1 http://arxiv.org/abs/2604.05678v1 Karim Bounja, Boujemaâ Achchab, Abdeljalil Sakat We introduce a natural displacement control for minimizer sets of oracle objectives equipped with certified epigraphic information. Formally, we replace the usual local uniform value control of objective perturbations - uncertifiable from finite pointwise information without additional structure - by the strictly weaker requirement of a cylinder-localized vertical epigraphic control, naturally provided by certified envelopes. Under set-based quadratic growth (allowing nonunique minimizers), this Tue, 07 Apr 2026 00:00:00 +0000 math.OC general-or arXiv http://arxiv.org/abs/2604.05640v1 http://arxiv.org/abs/2604.05640v1 Renzi Wang, Panagiotis Patrinos, Alberto Bemporad This paper presents a novel learning-based approach to construct a surrogate problem that approximates a given parametric nonconvex optimization problem. The surrogate function is designed to be the minimum of a finite set of functions, given by the composition of convex and monotonic terms, so that the surrogate problem can be solved directly through parallel convex optimization. As a proof of concept, numerical experiments on a nonconvex path tracking problem confirm the approximation quality Tue, 07 Apr 2026 00:00:00 +0000 math.OC general-or arXiv http://arxiv.org/abs/2604.05633v1 http://arxiv.org/abs/2604.05633v1 Yicheng Lin, Bingxian Wu, Nan Bai, Yunxiao Ren, Zhongkui Li et al. The Koopman operator enables simplified representations for nonlinear systems in data-driven optimal control, but the accompanying uncertainties inevitably induce deviations in the optimal controller and associated value function. This raises a distinct and fundamental question on optimality robustness, specifically, how uncertainties affect the optimal solution itself. To address this problem, we adopt a unified analysis-to-design perspective for systematically quantifying and improving optimal Tue, 07 Apr 2026 00:00:00 +0000 eess.SY general-or arXiv http://arxiv.org/abs/2604.05626v1 http://arxiv.org/abs/2604.05626v1 Pedro Aceves-Sanchez, Giacomo Albi, Federica Ferrarese, Michael Herty In this paper, we introduce a novel variant of the CBO method that incorporates jumps according to an $α$-stable stochastic process in a kinetic framework. This extension gives rise to nonlocal stochastic effects, which improve the exploration capabilities of the method. We formulate the method at the particle level, detailing the corresponding stochastic dynamics and its asymptotic behavior. In particular, through a Fourier-based representation, we derive the associated fractional Fokker-Planck Tue, 07 Apr 2026 00:00:00 +0000 math.OC general-or arXiv http://arxiv.org/abs/2604.05587v1 http://arxiv.org/abs/2604.05587v1 Zhe Zhao, Haibin Wen, Jiaming Ma, Jiachang Zhan, Tianyi Xu et al. An important recurring pattern in scientific breakthroughs is a two-stage process: an initial phase of undirected experimentation that yields an unexpected finding, followed by a retrospective phase that explains why the finding works and situates it within existing theory. We present ResearchEVO, an end-to-end framework that computationally instantiates this discover-then-explain paradigm. The Evolution Phase employs LLM-guided bi-dimensional co-evolution -- simultaneously optimizing both algor Tue, 07 Apr 2026 00:00:00 +0000 cs.AI stochastic ml-for-or arXiv http://arxiv.org/abs/2604.05567v1 http://arxiv.org/abs/2604.05567v1 Eder Baron-Prada, Julius P. J. Krebbekx, Adolfo Anta, Florian Dörfler Scaled graphs (SGs) offer a geometric framework for feedback stability analysis. This paper develops containment conditions for SGs within multiplier-defined regions, addressing both circular and conic geometries. For circular regions, we show that soft and hard SG containment are equivalent whenever the associated multiplier is positive-negative. This enables hard stability certification from soft computations alone, bypassing both the positive semidefinite storage constraint and the homotopy c Tue, 07 Apr 2026 00:00:00 +0000 math.OC general-or arXiv http://arxiv.org/abs/2604.05566v1 http://arxiv.org/abs/2604.05566v1 Perceval Beja-Battais, Guillaume Dupré, Alain Grossetête, Nicolas Vayatis Driven by advances in hardware and software technologies, nonlinear model predictive control (NMPC) has gained increasing adoption in both industry and academia over the past decades. However, its practical deployment is often limited by the computational cost of simulating the embedded process model, especially for high-dimensional, multi-time-scale, or nonlinear systems commonly found in real-world applications. Thus, this paper introduces Surrogate Dynamics Optimization (SDO), a warm-start fr Tue, 07 Apr 2026 00:00:00 +0000 math.OC ml-for-or arXiv http://arxiv.org/abs/2604.05518v1 http://arxiv.org/abs/2604.05518v1 Kaito Ito, Alexandre Proutiere We propose an active learning algorithm for linear system identification with optimal centered noise excitation. Notably, our algorithm, based on ordinary least squares and semidefinite programming, attains the minimal sample complexity while allowing for efficient computation of an estimate of a system matrix. More specifically, we first establish lower bounds of the sample complexity for any active learning algorithm to attain the prescribed accuracy and confidence levels. Next, we derive a sa Tue, 07 Apr 2026 00:00:00 +0000 math.OC general-or arXiv http://arxiv.org/abs/2604.05511v1 http://arxiv.org/abs/2604.05511v1 Michel Fliess, Claude Lobry, Emmanuel Trélat We revisit finite-dimensional linear-quadratic optimal control from the viewpoint of differential flatness. If the pair (A, B) is controllable, then the linear control system is flat, and every trajectory can be parametrized by a flat output and finitely many of its derivatives. Once this parametrization is inserted into the quadratic functional, the Euler-Lagrange condition becomes a linear differential equation with constant coefficients, or more generally a polynomial matrix differential equa Tue, 07 Apr 2026 00:00:00 +0000 math.OC general-or arXiv http://arxiv.org/abs/2604.05495v1 http://arxiv.org/abs/2604.05495v1 Michael T. M. Emmerich, Ksenia Pereverdieva, André H. Deutz The Solow--Polasky diversity indicator (or magnitude) is a classical measure of diversity based on pairwise distances. It has applications in ecology, conservation planning, and, more recently, in algorithmic subset selection and diversity optimization. In this note, we investigate the computational complexity of selecting a subset of fixed cardinality from a finite set so as to maximize the Solow--Polasky diversity value. We prove that this problem is NP-hard in general metric spaces. The reduc Tue, 07 Apr 2026 00:00:00 +0000 cs.CG general-or arXiv http://arxiv.org/abs/2604.05443v1 http://arxiv.org/abs/2604.05443v1 Ruixue Li, Wenjing Yang, Zhaorong Zhang, Xun Li, Juanjuan Xu In this paper, we investigate the distributed optimal control problem for a kind of nonlinear multi-agent systems. In particular,both the state and the system dynamic structures of each agent are private and can only be shared among communicating agents.This type of information structure is inevitable in fields such as collaborative control for industrial confidentiality, and renders traditional distributed control methods using all systems' dynamic structures ineffective. The primary contributi Tue, 07 Apr 2026 00:00:00 +0000 math.OC general-or arXiv http://arxiv.org/abs/2604.05398v1 http://arxiv.org/abs/2604.05398v1 Liya Guo, Ruimeng Hu, Xu Yang, Yi Zhu Continuous-time stochastic control with time-inhomogeneous jump-diffusion dynamics is central in finance and economics, but computing optimal policies is difficult under explicit time dependence, discontinuous shocks, and high dimensionality. We propose an actor-critic framework that serves as a mesh-free solver for entropy-regularized control problems and stochastic games with jumps. The approach is built on a time-inhomogeneous little q-function and an appropriate occupation measure, yielding Tue, 07 Apr 2026 00:00:00 +0000 math.OC general-or arXiv http://arxiv.org/abs/2604.05390v1 http://arxiv.org/abs/2604.05390v1 Wenjing Yang, Zhaorong Zhang, Xun Li, Juanjuan Xu In this paper, we investigate the distributed load frequency control problem in a multi-area smart grid under external load disturbances and measurement noise. The novelty lies in that the information privacy is fully taken into account, that is, the internal structural parameters and operational states of each area are not shared with non-neighboring areas, which makes traditional distributed optimal control methods ineffective. The main contribution is to propose a distributed algorithm for th Tue, 07 Apr 2026 00:00:00 +0000 math.OC general-or arXiv http://arxiv.org/abs/2604.05372v1 http://arxiv.org/abs/2604.05372v1 Hyeontae Jo Many optimization problems in science and engineering involve objective functions that are nonsmooth at their minimizers. A common strategy is to trace a branch of minimizers of a regularized objective as the smoothing scale tends to zero; however, for nonsmooth functions, it is generally unclear whether such a branch can be continued and whether the associated continuation equation remains locally solvable. We study heat-kernel regularization and the resulting continuation equation along a loca Tue, 07 Apr 2026 00:00:00 +0000 math.OC general-or arXiv http://arxiv.org/abs/2604.05294v1 http://arxiv.org/abs/2604.05294v1 Tao Zhang, Peter E. Caines This paper investigates mean field games and control on sparse networks. In the case of large expander graphs, the limit topologies are analyzed using the graphexon framework, which characterizes both dense network limits and sparse connections. We prove that the sequence of empirical graphexon measures defined on finite graphs converges weakly to a limit graphexon measure on a continuous state space. Furthermore, the associated sequence of discrete averaging operators converges strongly to a co Tue, 07 Apr 2026 00:00:00 +0000 math.OC general-or arXiv http://arxiv.org/abs/2604.05269v1 http://arxiv.org/abs/2604.05269v1 Nour Al Dandachly, Shuang Gao, Roland Malhamé This paper addresses the decentralized coordinated charging problem for a large population of battery storage agents (e.g. residential batteries, electrical vehicles, charging station batteries) using Mean Field Game (MFG). Agents are assumed to have affine dynamics and are coupled through a price that is continuous and monotonically increasing with respect to the difference between the average charging power and the grid's desired average charging power. An important modeling feature of the pro Tue, 07 Apr 2026 00:00:00 +0000 eess.SY general-or arXiv http://arxiv.org/abs/2604.06169v1 http://arxiv.org/abs/2604.06169v1 Guhao Feng, Shengjie Luo, Kai Hua, Ge Zhang, Di He et al. The static ``train then deploy" paradigm fundamentally limits Large Language Models (LLMs) from dynamically adapting their weights in response to continuous streams of new information inherent in real-world tasks. Test-Time Training (TTT) offers a compelling alternative by updating a subset of model parameters (fast weights) at inference time, yet its potential in the current LLM ecosystem is hindered by critical barriers including architectural incompatibility, computational inefficiency and mi Tue, 07 Apr 2026 00:00:00 +0000 cs.LG general-or arXiv http://arxiv.org/abs/2604.06161v1 http://arxiv.org/abs/2604.06161v1 Zhengming Yu, Li Ma, Mingming He, Leo Isikdogan, Yuancheng Xu et al. Most digital videos are stored in 8-bit low dynamic range (LDR) formats, where much of the original high dynamic range (HDR) scene radiance is lost due to saturation and quantization. This loss of highlight and shadow detail precludes mapping accurate luminance to HDR displays and limits meaningful re-exposure in post-production workflows. Although techniques have been proposed to convert LDR images to HDR through dynamic range expansion, they struggle to restore realistic detail in the over- an Tue, 07 Apr 2026 00:00:00 +0000 cs.CV survey arXiv http://arxiv.org/abs/2604.06156v1 http://arxiv.org/abs/2604.06156v1 Yuchi Wang, Haiyang Yu, Weikang Bian, Jiefeng Long, Xiao Liang et al. MLLMs have been successfully applied to multimodal embedding tasks, yet their generative reasoning capabilities remain underutilized. Directly incorporating chain-of-thought reasoning into embedding learning introduces two fundamental challenges. First, structural misalignment between instance-level reasoning and pairwise contrastive supervision may lead to shortcut behavior, where the model merely learns the superficial format of reasoning. Second, reasoning is not universally beneficial for em Tue, 07 Apr 2026 00:00:00 +0000 cs.CV ml-for-or survey arXiv http://arxiv.org/abs/2604.06155v1 http://arxiv.org/abs/2604.06155v1 Qimin Zhong, Hao Liao, Haiming Qin, Mingyang Zhou, Rui Mao et al. Whether Large Language Models (LLMs) develop coherent internal world models remains a core debate. While conventional Next-Token Prediction (NTP) focuses on one-step-ahead supervision, Multi-Token Prediction (MTP) has shown promise in learning more structured representations. In this work, we provide a theoretical perspective analyzing the gradient inductive bias of MTP, supported by empirical evidence, showing that MTP promotes the convergence toward internal belief states by inducing represent Tue, 07 Apr 2026 00:00:00 +0000 cs.LG general-or arXiv http://arxiv.org/abs/2604.06148v1 http://arxiv.org/abs/2604.06148v1 Andrew Kurtz, Klaudia Krawiecka The governance of artificial intelligence has a blind spot: the machine identities that AI systems use to act. AI agents, service accounts, API tokens, and automated workflows now outnumber human identities in enterprise environments by ratios exceeding 80 to 1, yet no integrated framework exists to govern them. A single ungoverned automated agent produced $5.4-10 billion in losses in the 2024 CrowdStrike outage; nation-state actors including Silk Typhoon and Salt Typhoon have operationalized un Tue, 07 Apr 2026 00:00:00 +0000 cs.CR general-or arXiv http://arxiv.org/abs/2604.06138v1 http://arxiv.org/abs/2604.06138v1 Yanis Labrak, David Grünert, Séverin Baroudi, Jiyun Chun, Pawel Cyrta et al. Long-context audio reasoning is underserved in both training data and evaluation. Existing benchmarks target short-context tasks, and the open-ended generation tasks most relevant to long-context reasoning pose well-known challenges for automatic evaluation. We propose a synthetic data generation pipeline designed to serve both as a training resource and as a controlled evaluation environment, and instantiate it for first-visit doctor-patient conversations with SOAP note generation as the task. Tue, 07 Apr 2026 00:00:00 +0000 cs.SD healthcare-or arXiv