arXiv:2606.26712v1 Announce Type: cross Abstract: Skin lesion segmentation is a key task in computer-aided dermatological diagnosis, where accuracy directly impacts downstream analysis and disease classification. However, dermoscopic images are challenging due to blurred boundaries, low contrast, large shape variations, and artifacts such as hair and shadows. Recently, diffusion models have shown strong performance in medical image segmentation thanks to their progressive denoising and distribution modeling capabilities. Nevertheless, existing diffusion-based methods still suffer from limited cross-level feature interaction and insufficient boundary detail recovery. To address these issues, we propose MLFFM-SegDiff, a multi-level feature fusion diffusion model for skin lesion segmentation. Built on a diffusion framework, the method introduces a dual-path U-Net encoder, a Multi-Level Feature Fusion Module (MLFFM), and a boundary-sensitive loss function. The dual-path encoder enhances interaction between noisy mask features and dermoscopic image features. MLFFM improves skip connections via attention, scale alignment, and adaptive cross-level fusion. These designs enable the decoder to jointly leverage shallow boundary cues and deep semantic representations, improving mask reconstruction quality. Experiments on ISIC2018, PH2, and HAM10000 demonstrate that MLFFM-SegDiff outperforms representative methods including DermoSegDiff, U-Net, and SwinUNETR across Accuracy, F1-score, Jaccard index, Recall, and Dice. In particular, it achieves an average Jaccard index of 0.8546 and Dice coefficient of 0.9207. These results validate the effectiveness of the proposed multi-level feature fusion strategy for improving lesion segmentation performance. The code will be released at https://github.com/Qacket/MLFFM-SegDiff.git after publication.
Science Journals
arXiv:2606.27114v1 Announce Type: new Abstract: Uplift modeling, crucial for estimating individual treatment effects (ITE), faces dual challenges: flexibly leveraging inter-group similarity to enhance discriminative power and debiasing under unobserved confounding scenarios. In this paper, we propose the Cross-Head Attention Uplift Network (CHAUN) and Robust Adversarial Inverse Propensity Score (RA-IPS) method to address these limitations. CHAUN employs shared feature embeddings and cross-head attention mechanisms to dynamically integrate treatment-specific and control-specific representations, enhancing inter-group correlation modeling. Theoretically, we prove that access to the true propensity scores ensures ITE identifiability even with unobserved confounders. For practical scenarios lacking true propensity scores, RA-IPS adversarially optimizes propensity weights within constrained uncertainty sets to mitigate bias from unobserved variables. Experiments on public datasets (CRITEO-UPLIFT, LAZADA) and a production e-commerce dataset demonstrate CHAUN's superiority over state-of-the-art uplift models, achieving relative improvements of up to 25.6% in QINI scores. RA-IPS further enhances robustness, outperforming standard IPS by 5.4% under unobserved confounding. The results validate the effectiveness of our proposed methods in real-world causal inference tasks.
arXiv:2606.27122v1 Announce Type: new Abstract: Translating C programs to safe Rust is challenging owing to significant differences in typing constraints, ownership, and borrowing rules. Interpreter programs are particularly important targets for such translation, as they often handle untrusted inputs and suffer from memory-related vulnerabilities. We present Reboot, a mostly-automatic technique that translates real-world interpreter programs from C to safe Rust. Using Reboot, we have translated six interpreters ranging from 6k to 23k lines of C code to safe Rust, with each translation requiring only 1 to 11 brief user interventions. All translations pass 100% of the provided test suites, and achieve 62%--92% pass rates on separately created validation tests that were never exposed to the system. A security case study on mujs shows that memory vulnerabilities such as heap buffer overflows and use-after-free present in C are eliminated in the safe Rust translation. Two ideas underpin Reboot. First, feature reduction decomposes the translation by program features, creating a sequence of milestones where each is a complete, testable program; the translation starts from the simplest version and incrementally restores features, with each milestone validated before proceeding. Second, a multi-agent architecture orchestrates inherently unreliable coding agents through automated validation and feedback, keeping long-running translation workflows on track with minimal human involvement. An ablation study confirms that feature reduction improves translation correctness compared to using multi-agent translation alone, with 6%--20% improvements in pass rates on validation test suites.
arXiv:2606.26991v1 Announce Type: cross Abstract: X-ray computed tomography reconstruction is an ill-posed inverse problem, particularly in low-dose and sparse-angle settings where measurements are noisy and incomplete. While learned reconstruction methods such as the Learned Primal-Dual algorithm achieve strong performance, they typically rely on supervised training with access to ground-truth data, which is often unavailable in practice. In this work, we propose a self-supervised reconstruction method by extending the Noise2Inverse framework to the Learned Primal-Dual algorithm. The resulting approach, called Noise2Inverse Learned Primal-Dual (N2I-LPD), enables training of a learned iterative reconstruction operator without ground-truth images by exploiting the statistical independence of noise in distinct measurements with respect to angular rotation of the CT-scan. We compare the proposed method with classical reconstruction methods, as well as neural network-based approaches such as a U-Net trained within the same N2I framework. The results demonstrate that N2I-LPD achieves improved reconstruction quality, highlighting the potential of combining learned reconstruction operators with self-supervised training strategies for practical CT imaging scenarios where ground-truth data is unavailable.
arXiv:2606.27119v1 Announce Type: cross Abstract: Foundation decoders, a class of high-capacity neural decoders, are leading candidates for fault-tolerant quantum computing, with accurate and efficient decoding at large code distances. However, their construction often faces a steep scaling barrier, as larger code distances rapidly amplify the cost of syndrome generation and neural optimization. To address this bottleneck, here we devise neural transfer unification (NTU), a unified framework for efficient foundation decoders. A central feature of NTU is its ability to align decoding tasks across code distances via algebraic structures shared by scalable code families, which enables knowledge learned on smaller codes to accelerate large-scale decoder training. We instantiate NTU as NTU-Transformer, a transformer-based neural decoder tailored for planar surface codes and bivariate bicycle codes. For planar surface codes under circuit-level noise, NTU-Transformer outperforms correlation-aware matching on the $[\![361,1,19]\!]$ code and further scales to the $[\![625,1,25]\!]$ code, where it exceeds standard matching through transfer adaptation. For the bivariate bicycle code with $[\![72,12,6]\!]$, it surpasses Relay-BP in the low-physical-error regime. These results establish our proposal as a scalable route to amortized cross-distance training of foundation decoders for fault-tolerant quantum processors.
arXiv:2606.27133v1 Announce Type: cross Abstract: Understanding the dynamic behavior of complex biomolecules requires simplified models that not only make computations feasible but also reveal fundamental mechanisms. Coarse-graining (CG) achieves this by grouping atoms into beads, whose stochastic dynamics can be derived using the Mori-Zwanzig formalism, capturing both reversible and irreversible interactions. In liquid, the dissipative bead-bead interactions have so far been restricted to hydrodynamic couplings. However, friction does not only arises from the solvent but notably, from the internal degrees of freedom missing in the CG beads. This leads to an additional ''internal friction'' whose relevance is studied in this contribution. By comparing with all-atom molecular dynamics (MD), we neatly show that in order to accurately reproduce the dynamics of a globular protein in water using a coarse-grained (CG) model, not only a precise determination of elastic couplings and the Stokesian self-friction of each bead is required. Critically, the inclusion of internal friction between beads is also necessary for a faithful representation of protein dynamics. We propose to optimize the parameters of the CG model through a self-averaging method that integrates the CG dynamics with an evolution equation for the CG parameters. This approach ensures that selected quantities, such as the radial distribution function and the time correlation of bead velocities, match the corresponding MD values.
arXiv:2606.27159v1 Announce Type: cross Abstract: For a graph $G$, a proper $k$-coloring of $G$ is \emph{equitable} if the sizes of any two color classes differ by at most one. The \textsc{Equitable $k$-Coloring} problem asks, for a given graph $G$ and integer $k$, whether $G$ admits an equitable $k$-coloring. Bodlaender and Fomin showed that it is polynomial-time solvable on graphs of bounded treewidth, while it remains $\NP$-hard on cographs, and thus on graphs of constant clique-width. Fellows et al. showed that the problem becomes $\mathsf{W[1]}$-hard when parameterized by tree-width (and hence clique-width) plus the number of colors~$k$. We first show that, for every fixed $k$, counting equitable $k$-colorings is polynomial-time solvable on graph classes of bounded clique-width, given a clique-width expression. We then show that, under $\mathsf{SETH}$, the dependence on clique-width in this algorithm is essentially optimal. As a consequence, our results provide a fairly tight picture of the complexity of \textsc{Equitable $k$-Coloring} with respect to the combined parameter $k$+clique-width. Second, we refine our clique-width algorithm for the linear setting. We show that there exists an algorithm, given an integer $k\ge 1$ and an $n$-vertex graph $G$ together with a linear $w$-expression constructing $G$, computes the number of equitable $k$-colorings of $G$ in time $\max\{1,2^k-2\}^w\cdot n^{k+O(1)}$. Third, we consider a different structural restriction, namely the class of $P_t$-free graphs. A graph is called $P_t$-free if it does not contain the path on $t$ vertices as an induced subgraph. This is a different setting from bounded clique-width; in particular, already $P_5$-free graphs have unbounded clique-width. Nevertheless, we show that for every $P_t$-free graph $G$, the number of equitable list $3$-colorings of $G$ can be computed in subexponential time.
arXiv:2606.27335v1 Announce Type: cross Abstract: A flexible forward (FF) is a customized FX hedging instrument that guarantees a fixed exchange rate while letting the holder choose the delivery date within a pre-agreed window. It is therefore an American-style option on timing, and its valuation must respect the volatility skew of the underlying currency pair. We price FF contracts (and, more generally, American options) under a time-inhomogeneous Heston model which captures the forward-skew term structure while preserving analytical tractability through a recursive (matrix) Riccati solution for the joint characteristic function. Extending the integral-equation (decomposition) approach to time-dependent coefficients, we derive a Volterra equation characterizing the early-exercise surface. The expectation in the decomposition formula is evaluated by two complementary spectral methods: a double cosine (COS) expansion of the transition density, and a damped-Sinc (DSINC) local-basis scheme that is more accurate and stays robust when a low Feller ratio or large vol-of-vol induces Gibbs oscillations in the COS series. Benchmarked against a penalty-iteration MCS-ADI finite-difference solver, both methods price a contract in about 1-2 seconds, roughly an order of magnitude faster than the finest finite-difference grid, while DSINC improves median accuracy over COS by about a factor of twelve. The experiments also show that the early-exercise surface is a substantially nonlinear function of the variance, contrary to the linear-in-variance approximation common in earlier work.
arXiv:2308.11033v2 Announce Type: replace Abstract: Modern society relies heavily on infrastructure networks, from communication to power systems, making their reliability under failure and disruption critical. Robustness typically requires redundancy, providing alternative paths that ensure continuous service, yet redundancy is costly and often limited. Real-world networks therefore operate under a fundamental trade-off between resource investment and reliability. Here we derive theoretical bounds governing this trade-off and identify the network architectures that minimize expected downtime under constraints of cost, component durability, and spatial embedding. We show that, in the relevant limit, optimal configurations consist of a 3-connected structural core linked by intermediate chains of uniform weight. This reveals a previously unrecognized family of spatial networks that achieves near-optimal robustness with orders-of-magnitude fewer resources. Distinct from commonly studied network models, these architectures expose general structural principles underlying the efficient organization of reliable infrastructure networks.
arXiv:2404.18054v2 Announce Type: replace Abstract: As one kind of important phase field equations, Cahn-Hilliard equations involve high-order spatial derivatives, strong nonlinearities, and even solution singularities when certain bulk potentials are used. When using the physics informed neural network (PINN) to simulate the long time evolution of the solution, it is necessary to decompose the time domain to capture the transition of solutions in different time. Moreover, the standard PINN cannot maintain the mass conservation property for the equations exactly. We propose a novel mass-preserving spatiotemporal adaptive PINN, which adaptively divides the time domain according to the rate of energy decrease, and solves the Cahn-Hilliard equation within each subinterval. To improve the prediction accuracy, spatial adaptive sampling is employed in the subdomain to select points with large residual value which are added to the training samples. Notably, a mass constraint is added to the loss function to compensate the mass degradation problem of the PINN method when solving Cahn-Hilliard equations. Numerical experiments are presented to illustrate the effectiveness of the proposed method in solving complex phase field models, including the Cahn-Hilliard equations with different bulk potentials, the three-dimensional Cahn-Hilliard equation with singularities, and the system of Cahn-Hilliard equations.
arXiv:2405.11454v3 Announce Type: replace Abstract: We study gradient testing and gradient estimation of smooth functions using only a comparison oracle that, given two points, indicates which one has the larger function value. For any smooth $f\colon\mathbb R^n\to\mathbb R$, $\mathbf{x}\in\mathbb R^n$, and $\varepsilon>0$, we design a gradient testing algorithm that determines whether the normalized gradient $\nabla f(\mathbf{x})/\|\nabla f(\mathbf{x})\|$ is $\varepsilon$-close or $2\varepsilon$-far from a given unit vector $\mathbf{v}$ using $O(1)$ queries, as well as a gradient estimation algorithm that outputs an $\varepsilon$-estimate of $\nabla f(\mathbf{x})/\|\nabla f(\mathbf{x})\|$ using $O(n\log(1/\varepsilon))$ queries which we prove to be optimal. Furthermore, we study gradient estimation in the quantum comparison oracle model where queries can be made in superpositions, and develop a quantum algorithm using $O(\log (n/\varepsilon))$ queries.
arXiv:2406.07147v3 Announce Type: replace Abstract: This study employs cutting-edge wearable monitoring technology to conduct high-precision, high-temporal-resolution (1-second interval) cognitive load assessment on electroencephalogram (EEG) data from the FP1 channel and heart rate variability (HRV) data of secondary vocational students. By jointly analyzing these two critical physiological indicators, the research delves into their application value in assessing cognitive load among secondary vocational students and their utility across various tasks. The study designed two experiments to validate the efficacy of the proposed approach: Initially, a random forest classification model, developed using the N-BACK task, enabled the precise decoding of physiological signal characteristics in secondary vocational students under different levels of cognitive load, achieving a classification accuracy of 97%. Subsequently, this classification model was applied in a cross-task experiment involving the National Computer Rank Examination (Level-1), demonstrating the method's significant applicability and cross-task transferability in diverse learning contexts. Conducted with high portability, this research holds substantial theoretical and practical significance for optimizing teaching resource allocation in secondary vocational education, as well as for cognitive load assessment methods and monitoring. Currently, the research findings are undergoing trial implementation in the school.
arXiv:2409.01447v3 Announce Type: replace Abstract: We present a finite-sample analysis of decentralized learning in two-player zero-sum matrix games and stochastic games, with a focus on best-response-based learning algorithms. In matrix games, the learning algorithm is payoff-based and symmetric: each player updates its policy using only its own payoff observations, incrementally moving toward an estimated smoothed best response to the opponent's latest policy. For stochastic games, we build on this matrix-game primitive to develop a learning algorithm called value iteration with smoothed best response (VI-SBR), which combines smoothed-best-response learning in induced matrix games with a decentralized, model-free approximation of minimax value iteration. We establish finite-sample guarantees in both settings. For matrix games, our results imply a sample complexity of $\mathcal{O}(\epsilon^{-1})$ for finding an $\epsilon$-Nash distribution and, with explicit exploration, $\tilde{\mathcal{O}}(\epsilon^{-8})$ for finding an $\epsilon$-Nash equilibrium. For stochastic games, we prove that the exploration-enhanced VI-SBR algorithm achieves a sample complexity of $\tilde{\mathcal{O}}(\epsilon^{-8})$ for finding an $\epsilon$-Nash equilibrium. Technically, our analysis develops a coupled Lyapunov-drift framework. This framework simultaneously handles stochastic iterative algorithms with multiple interacting stochastic iterates, the non-zero-sum auxiliary games generated by independently updated value functions, and the time-inhomogeneous Markovian noise induced by time-varying policies. The resulting tools may be useful more broadly for analyzing learning algorithms with coupled stochastic iterates and nonstationary sampling processes.
Communication-Efficient, 2D Parallel Stochastic Gradient Descent for Distributed-Memory Optimization
arXiv:2501.07526v2 Announce Type: replace Abstract: Distributed-memory implementations of numerical optimization algorithm, such as stochastic gradient descent (SGD), require interprocessor communication at every iteration of the algorithm. On modern distributed-memory clusters where communication is more expensive than computation, the scalability and performance of these algorithms are limited by communication cost. This work generalizes prior work on 1D $s$-step SGD and 1D Federated SGD with Averaging (FedAvg) to yield a 2D parallel SGD method (HybridSGD) which attains a continuous performance trade off between the two baseline algorithms. We present theoretical analysis which show the convergence, computation, communication, and memory trade offs between $s$-step SGD, FedAvg, 2D parallel SGD, and other parallel SGD variants. We implement all algorithms in C++ and MPI and evaluate their performance on a Cray EX supercomputing system. Our empirical results show that HybridSGD achieves better convergence than FedAvg at similar processor scales while attaining speedups of $5.3\times$ over $s$-step SGD and speedups up to $121\times$ over FedAvg when used to solve binary classification tasks using the convex, logistic regression model on datasets obtained from the LIBSVM repository.
arXiv:2503.15283v2 Announce Type: replace Abstract: Text-and-Image-To-Image (TI2I), an extension of Text-To-Image (T2I), integrates image inputs with textual instructions to enhance image generation. Existing methods often partially utilize image inputs, focusing on specific elements like objects or styles, or they experience a decline in generation quality with complex, multi-image instructions. To overcome these challenges, we introduce Training-Free Text-and-Image-to-Image (TF-TI2I), which adapts cutting-edge T2I models such as SD3 without the need for additional training. Our method capitalizes on the MM-DiT architecture, in which we point out that textual tokens can implicitly learn visual information from vision tokens. We enhance this interaction by extracting a condensed visual representation from reference images, facilitating selective information sharing through Reference Contextual Masking -- this technique confines the usage of contextual tokens to instruction-relevant visual information. Additionally, our Winner-Takes-All module mitigates distribution shifts by prioritizing the most pertinent references for each vision token. Addressing the gap in TI2I evaluation, we also introduce the FG-TI2I Bench, a comprehensive benchmark tailored for TI2I and compatible with existing T2I methods. Our approach shows robust performance across various benchmarks, confirming its effectiveness in handling complex image-generation tasks.
arXiv:2505.02870v3 Announce Type: replace Abstract: We propose a structural variational resolution of the Abraham-Lorentz-Dirac (ALD) pathologies. By deriving the Variational Kinematic Constraint (VKC) and the Variational Dynamics Constraint (VDC) from the particle's proper-time perspective, we show that self-induced variations are forbidden and dynamics arise solely from first-order proper-time variations of external fields. Consequently, self-force terms are excluded at the variational level, eliminating runaway solutions and non-causal behavior without regularization. Our framework further provides a first-principles derivation of minimal coupling and reveals gauge invariance as a necessary consequence of proper-time-based variational structure.
arXiv:2505.19922v4 Announce Type: replace Abstract: This paper describes the developed setup and characterization approach for full two-port calibrated S-parameter measurements at cryogenic temperatures, together with a complete uncertainty budget. The system developed at the Istituto Nazionale di Ricerca Metrologica (INRiM, Italy), exploits the Short-Open-Load-Reciprocal technique to realize error-corrected cryogenic measurements within a single cooling cycle. The system operates down to mK temperatures over the 4-12 GHz band in coaxial line. Calibration standards are referenced to SI-traceable room-temperature measurements, while a numerical approach is used to evaluate the response shift of the artifacts upon cooling and to derive an additional cryogenic uncertainty contribution for the measurement uncertainty budget. Moreover, relevant measurement uncertainty contributions are evaluated according to internationally agreed procedures, and a comprehensive uncertainty budget is presented. Test measurements on a 20 dB attenuator are shown as an example. An attenuation of 20.70 +/- 0.08 dB (95% confidence interval) was obtained at 6 GHz. Full SI-traceable verification at mK temperatures remains an open challenge; however, an initial calibration verification is also presented.
arXiv:2505.21581v4 Announce Type: replace Abstract: While end-to-end autonomous driving has advanced significantly, prevailing methods remain fundamentally misaligned with human cognitive principles in both perception and planning. In this paper, we propose CogAD, a novel end-to-end autonomous driving model that emulates the hierarchical cognition mechanisms of human drivers. CogAD implements dual hierarchical mechanisms: global-to-local context processing for human-like perception and intent-conditioned multi-mode trajectory generation for cognitively-inspired planning. The proposed method demonstrates three principal advantages: comprehensive environmental understanding through hierarchical perception, robust planning exploration enabled by multi-level planning, and diverse yet reasonable multi-modal trajectory generation facilitated by dual-level uncertainty modeling. Extensive experiments on nuScenes and Bench2Drive demonstrate that CogAD achieves state-of-the-art performance in end-to-end planning, exhibiting particular superiority in long-tail scenarios and robust generalization to complex real-world driving conditions.
arXiv:2505.23527v4 Announce Type: replace Abstract: Modern reinforcement learning (RL) algorithms have found success by using powerful probabilistic models, such as transformers, energy-based models, and diffusion/flow-based models. To this end, RL researchers often choose to pay the price of accommodating these models into their algorithms -- diffusion models are expressive, but are computationally intensive due to their reliance on solving differential equations, while autoregressive transformer models are scalable but typically require learning discrete representations. Normalizing flows (NFs), by contrast, seem to provide an appealing alternative, as they enable likelihoods and sampling without solving differential equations or autoregressive architectures. However, their potential in RL has received limited attention, partly due to the prevailing belief that normalizing flows lack sufficient expressivity. We show that this is not the case. Building on recent work in NFs, we propose a single NF architecture which integrates seamlessly into RL algorithms, serving as a policy, Q-function, and occupancy measure. Our approach leads to much simpler algorithms, and achieves higher performance in imitation learning, offline, goal conditioned RL and unsupervised RL.
arXiv:2506.07031v5 Announce Type: replace Abstract: Emerging Large Reasoning Models (LRMs) consistently excel in mathematical and reasoning tasks, showcasing remarkable capabilities. However, the enhancement of reasoning abilities and the exposure of internal reasoning processes introduce new safety vulnerabilities. A critical question arises: when reasoning becomes intertwined with harmfulness, will LRMs become more vulnerable to jailbreaks in reasoning mode? To investigate this, we introduce HauntAttack, a novel and general-purpose black-box adversarial attack framework that systematically embeds harmful instructions into reasoning questions. Specifically, we modify key reasoning conditions in existing questions with harmful instructions, thereby constructing a reasoning pathway that guides the model step by step toward unsafe outputs. We evaluate HauntAttack on 11 LRMs and observe an average attack success rate of over 70\%, achieving up to 13 percentage points of absolute improvement over the strongest prior baseline. Our further analysis reveals that even advanced safety-aligned models remain highly susceptible to reasoning-based attacks, offering insights into the urgent challenge of balancing reasoning capability and safety in future model development.
arXiv:2508.03247v2 Announce Type: replace Abstract: Prior clinical psychology research shows that Western individuals with depression tend to report psychological symptoms, while Eastern individuals report somatic ones. We test whether Large Language Models (LLMs), which are increasingly used in mental health, reproduce these cultural patterns by prompting them with Western or Eastern personas. Results show that LLMs largely fail to replicate the patterns when prompted in English, though prompting in major Eastern languages (i.e., Chinese, Japanese, and Hindi) improves alignment in several configurations. Our analysis pinpoints two key reasons for this failure: the models' low sensitivity to cultural personas and a strong, culturally invariant symptom hierarchy that overrides cultural cues. These findings reveal that while prompt language is important, current general-purpose LLMs lack the robust, culture-aware capabilities essential for safe and effective mental health applications.
arXiv:2508.21221v2 Announce Type: replace Abstract: Lower limb exoskeletons show promise to assist human movement, but their utility is limited by controllers designed for discrete, predefined actions in controlled environments, restricting their real-world applicability. We present an uncertainty-aware control framework that enables ankle exoskeletons to operate safely across diverse scenarios by automatically disengaging when encountering unfamiliar movements. Our approach uses an uncertainty estimator to classify movements as similar (in-distribution) or different (out-of-distribution) relative to actions in the training set. We evaluated three architectures (model ensembles, autoencoders, and generative adversarial networks) on an offline dataset and tested the strongest performing architecture (ensemble of gait phase estimators) online. The online test demonstrated the ability of our uncertainty estimator to turn assistance on and off as the user transitioned between in-distribution and out-of-distribution tasks (F1: 89.2). This new framework provides a path for exoskeletons to safely and autonomously support human movement in unstructured, everyday environments.
arXiv:2509.15622v2 Announce Type: replace Abstract: Virtual analog (VA) modeling methods seek to emulate analog audio hardware using digital signal processing (DSP). Modeling approaches fall into three broad categories: white-box methods, which use detailed device knowledge for accurate simulation; gray-box methods that use generic DSP blocks to model the system; and black-box methods, which rely solely on opaque models learned from input-output data. A category of architectures used widely in black-box modeling are recurrent neural networks (RNNs). To model device controls, the control values can be provided as conditioning input to the network. However, when the conditioning is time-varied, the models are susceptible to producing noise artifacts. Regularization of the RNN dynamics significantly reduces these artifacts, though at a loss in modeling accuracy. This paper closes the dynamics regularization quality gap by introducing deep control-conditioned LSTMs and a gammatone filterband (GFB) loss. Experiments indicate that the proposed method achieves comparable modeling performance as unregularized baselines while avoiding the noise artifacts caused by time-varying control inputs.
arXiv:2510.15238v2 Announce Type: replace Abstract: Optimizing a single advertising campaign across heterogeneous channels is a central challenge in industrial autobidding. Auction mechanisms vary across channels in ranking rules (pure eCPM vs. UE-augmented scoring), pricing formats (first- vs. second-price), and bidding conventions (uniform vs. non-uniform), while advertisers impose shared campaign-level constraints. We propose HOB, which makes marginal cost (MC) computable and alignable across heterogeneous channels, especially for first-price auctions (FPA) with organic-paid coexistence, where existing bidding formulations do not yield a practical aligned MC form. At the global level, HOB derives channel-specific MC forms and coordinates disparate channels through a shared MC target. At the local level, HOB models free-win probability and winning-price uncertainty with a zero-inflated exponential distribution, yielding an efficient surplus-optimal bidding strategy for non-uniform first-price auctions. We show that any interior optimum satisfies MC equalization across channels. Experiments on a controlled offline benchmark, industrial log replay, and large-scale online A/B tests demonstrate that HOB consistently delivers significant performance gains. Deployed on a large-scale commercial DSP, HOB delivers a 3.0% lift in GMV while maintaining return on advertising spend (ROAS) constraints.
arXiv:2510.25629v3 Announce Type: replace Abstract: Pauli exchange-repulsion is the dominant short-range intermolecular interaction and it is an essential component of molecular force fields. Current approaches to modeling Pauli repulsion in molecular force fields often rely on over 20 atom types to achieve chemical accuracy, illustrating the challenge in finding models which have broadly transferable parameters, which hampers the development of force fields with quantum-chemical accuracy that are transferable across many chemical systems. We present the anisotropic valence density overlap (AVDO) model for exchange-repulsion. The model produces sub-kcal/mol accuracy for dimers of organic molecules and contains two universal parameters, which we demonstrate are transferable for molecules composed of H, C, N, O, F, P, S, Cl, and Br. The model is tested on 1,872 unique molecular pairs selected from a set of 135 molecules, and samples dissociation curves and configurations from condensed-phase molecular dynamics trajectories. Given recent progress in machine learning of the electronic density, this model offers a promising path toward high-accuracy, next-generation machine-learned force fields.