arXiv:2607.08620v1 Announce Type: new
Abstract: As human-robot interaction rapidly spreads in numerous fields, the subject of robot acceptance gains increasing importance. Visual similarity to the human body, as occurs for humanoids, is generally not enough to ensure acceptance in physical interaction, as acceptance directly links to comfort and ergonomics, which are measured in terms of the quality of the robot movement perceived by the human. This paper discusses the connection between comfort and similarity of the robot movement to the human one. By considering the kinematic characterization of human movement, this paper focuses on the time laws of such movements, wherein the end-effector path is prescribed. Based on the lognormality principle for modeling human movements, a human-likeness index is defined and used to provide an a priori characterization of trajectories. Such an index can be used to evaluate the performance of trajectory generation algorithms in producing human-like movements before they are actually executed. For validation purposes, 68 subjects are required to judge their comfort. The results of three experimental campaigns involving a physical interaction with a robot demonstrate a globally consistent trend between the preference in terms of perceived comfort and the distribution of the suggested human-likeness index.
Science Journals
arXiv:2607.08672v1 Announce Type: new
Abstract: Mechanisms of interaction in spreading models are central to our quantitative understanding of networked contagion processes, from disease transmission to opinion dynamics. Yet, while empirical data can reveal who interacts with whom, they rarely provide direct information about how interactions drive spreading, leaving the underlying mechanism to be inferred from observed dynamics, and selected among competing hypotheses. We propose a general framework for model selection in binary-state spreading processes on networks and show that asymptotic approximations in the thermodynamic limit can accurately predict inference outcomes in finite systems. By systematically exploring a broad parameter space, we characterize the detectability of six archetypal spreading mechanisms commonly used in the literature and find that accuracy generally increases in sparse networks, which are prevalent in real-world systems, and near phase transitions, such as the epidemic threshold of simple contagion processes. We further assess the prevalence of these mechanisms across a diverse set of empirical datasets, highlighting the impact of data preprocessing on model recovery. Our results show that statistical model selection can fail under common conditions and suggest new directions for overcoming these limitations.
arXiv:2607.08319v1 Announce Type: new
Abstract: We present GitLake, a Git-for-data design for an agent-first lakehouse. The system lifts single-table Iceberg snapshots into lakehouse-wide commits, branches, and merges, letting agents work on isolated branches while humans review and publish changes. Pipelines run on temporary branches and publish through a final merge, so all outputs become visible atomically or none do. Finally, we report production lessons as well as correctness insights from a preliminary Alloy model of our core abstractions.
arXiv:2607.08528v1 Announce Type: new
Abstract: This paper presents a modular method for generating reference signals online for saturable synchronous machine drives. The method dynamically generates optimal references without precomputed lookup tables, following the maximum-torque-per-ampere (MTPA) trajectory while respecting maximum-torque-per-volt (MTPV), current, and voltage limits. The proposed tracking laws are formulated to yield exact, decoupled first-order error dynamics, ensuring predictable tracking responses and simplifying system tuning. The algorithm requires only the forward flux map, thereby eliminating the need for current-map inversion. By operating in a feedforward manner, the method ensures noise-free reference signals and structural separation from the feedback control. Both simulation and experimental results are presented, demonstrating that the proposed method achieves dynamic and steady-state performance on par with conventional lookup-table-based approaches, while avoiding the need for precomputed reference tables.
arXiv:2607.08432v1 Announce Type: new
Abstract: A method is presented to carry out a transient simulation of eddy current problems with nonlinear materials. Coils are voltage-driven. The magnetic field due to currents in coils are considered by their Biot-Savart-fields. The magnetic vector potential is used in the finite element formulation. The time stepping method is based on implicit Euler. The arising nonlinear equation system is split into two parts, the common finite element system and a circuit equation. Each part is solved separately by Newton's method. Additionally, a line search is used to solve the nonlinear field equations. Inrush currents and average magnetic flux densities through cross sections of laminates of a nonlinear benchmark problem consisting of a laminated iron core inserted in a cylindrical coil are studied. All details of the numerical benchmark are given to evaluate the presented results. Numerical data describing the performance of the presented method are provided.
arXiv:2607.08539v1 Announce Type: new
Abstract: Leveraging large language models (LLMs) to analyze complex documents -- such as academic papers, technical manuals, and financial reports -- has emerged as a mainstream and critical task in both research and industry. In practice, users must first filter relevant documents from large collections and then conduct in-depth analysis (e.g. question answering) over the selected subset, yet existing systems flatten documents into plain-text chunks, discarding the rich hierarchical structures (sections, tables, figures, equations) and degrading downstream performance. We present DocMaster, a hierarchical structure-aware document analysis system. DocMaster parses documents into hierarchical document trees preserving original layouts and constructs a structure-aware semantic index that enables accurate document filtering and in-depth analysis. We demonstrate DocMaster through an interactive web interface that enables users to upload document collections, construct tree-based and multi-view semantic indices, filter relevant documents via natural-language conditions, and perform follow-up question answering over the filtered results. The source code, data, and demo are available at https://doc-master.github.io/.
arXiv:2607.08674v1 Announce Type: new
Abstract: The rapid advancement of generative AI has enabled the creation of highly realistic deepfake media, posing significant threats, including misinformation, digital identity theft, fraud, and manipulation of public opinion. AI-generated image (AIGI) detection is reliably challenging due to the diversity of generative methods and the subtle artifacts they leave behind. In this work, we propose GenRes, a novel framework for generative residual learning via a neural tensor network, which models fine-grained relational features between original and transformed samples to enhance generalization. To address scenarios involving multiple generative transformations, we introduce GenRes++, which employs a learnable attention mechanism to aggregate relational features across multiple transformed samples and enables the model to focus on the most informative cues. Both models leverage PE-Core as a feature extractor, providing generalized and semantically rich embeddings that improve cross-domain performance and enable the detection of AIGI generated by unseen methods. Comprehensive experiments on multiple benchmark datasets demonstrate that the proposed GenRes++ approach outperforms existing methods.
arXiv:2607.08681v1 Announce Type: new
Abstract: As agentic AI systems are increasingly applied to cyber-physical environments, their evaluation requires assessment of both task performance and trustworthiness. In decentralized energy markets, autonomous agents may improve market utility, but may also exploit invalid physical data, create artificial liquidity, and produce unstable governance decisions. Therefore, we propose SolarChain-Eval, a physics-constrained benchmark for evaluating trustworthy economic agents. It formulates market governance as a Gymnasium-compatible Markov Decision Process, where agents make hourly decisions. SolarChain-Eval evaluates each policy across multiple dimensions, including market utility, physical safety, slippage, action smoothness, spatial fairness, and auditability. To support agentic evaluation, SolarChain-Eval incorporates an LLM-based Planner/Auditor layer. The Planner defines episode-level action bounds and audit rules, while the Auditor reviews and revises high-risk actions. All interventions are recorded through structured logs, including trigger signals, proposed actions, revised actions, and audit rationales. Experiments with static, random, myopic, RL, and RL+LLM policies reveal a clear utility-safety trade-off. RL agents improve market utility but can still produce unsafe behavior. When the physics penalty is removed, reward-maximizing agents exploit invalid generation and increase artificial liquidity. The LLM Planner/Auditor improves auditability and mitigates selected risks, but it cannot fully compensate for a misspecified reward function. These results indicate that trustworthy agentic AI evaluation requires both physical constraints and transparent intervention traces. We release data and code as open access on GitHub for replicability.
arXiv:2607.08492v1 Announce Type: new
Abstract: We prove expression rate bounds of finite-parametric, spectral and neural surrogates for holomorphic maps between separable Hilbert spaces. The surrogates have an encoder-approximator-decoder architecture, with Karhunen-Lo\'{e}ve encoders and frame decoders. We prove expression rate bounds for two classes of finite-parametric surrogates: i) spectral surrogates obtained by N-term truncations of Wiener polynomial chaos expansions and ii) neural surrogates obtained by approximation of parametric maps with deep feedforward neural networks, ReLU and RePU activation functions and uniformly bounded weights. We work under an algebraic decay assumption on the eigenvalues of the covariance of the Gaussian measure on the input space. We obtain convergence rates for mean-square errors, and additionally in first-order Gaussian Sobolev spaces, to account for errors in the approximation of gradients.
arXiv:2607.08754v1 Announce Type: new
Abstract: Low-rank factorization is widely used to compress neural networks, but modern models are often not naturally amenable to aggressive factorization without significant accuracy loss. Existing training-time low-rank regularizers can improve compressibility, but they often require SVDs of large weight matrices, modify the model architecture (introducing additional trainable parameters), or rely on stateful cached quantities. To address these limitations, we introduce SLORR, a simple, stateless, and architecture-preserving framework for in-training low-rank regularization, instantiated with two main variants based on the Hoyer sparsity metric and the nuclear norm. SLORR directly regularizes the original weight matrices using GPU-friendly approximations for the forward and backward passes of the regularizers, for which we provide approximation guarantees. We first evaluate SLORR on ImageNet-1K across short-horizon continued training of ResNet-50, ViT-B/16, and ViT-L/16, and pretraining of ResNet-18, where SLORR induces compressibility while introducing less than 8% training overhead. We further evaluate SLORR-Hoyer in LLM pretraining at 135M and 560M scales: SLORR-trained compressed models preserve performance substantially better than unregularized models while adding less than 1% average training overhead.
arXiv:2607.07959v1 Announce Type: cross
Abstract: We study a mean-field reaction network whose species are assemblies built from identical atoms by reversible coagulation and fragmentation. Each assembly is an ordered binary tree, so the number of species of a given length grows combinatorially, as the Catalan numbers. The model nonetheless admits an explicit equilibrium and tractable stochastic dynamics. A finite volume $V$ sets a crossover length $l_c \sim \ln V$ that splits the equilibrium into two sectors. Below $l_c$ each assembly occurs in many copies and the rank-frequency distribution is Zipf-like; above $l_c$ individual species are rare and fluctuation-dominated. The statistical weight of the rare sector decays slowly with volume, controlling the finite-size scaling of diversity, Shannon entropy, and other assembly-weighted observables. The equilibrium also admits a transparent grand-canonical description in terms of a bond energy and an atomic chemical potential. Together these results make the model a controlled neutral baseline against which selection and driving in richer assembly networks can be measured.
arXiv:2607.07967v1 Announce Type: cross
Abstract: Diffusion-based policies have recently emerged as powerful policy parameterizations for reinforcement learning, representing state-conditioned action distributions as terminal laws of diffusion processes with parameterized drifts. This terminal-law representation has shown substantial expressive flexibility in practice, enabling diffusion policies to model complex, multimodal, and highly non-Gaussian action distributions; however, it remains unclear what mathematically drives this expressivity and how to fully exploit it when the policy is learned from finite data. In this paper, we identify the drift Lipschitz budget $K$ as a central quantity governing the expressivity and statistical behavior of diffusion policies. We quantify expressivity through approximation: diffusion policies with $K$-Lipschitz drifts can concentrate near optimal deterministic policies and achieve value approximation error of order $1/K$; moreover, we prove a matching lower bound under nondegenerate diffusion noise. This increased expressivity comes with a statistical cost. When the drift is parameterized by neural networks, increasing $K$ improves approximation but increases statistical complexity. Balancing these two terms yields a finite-sample performance gap of order $\tilde{O}(n^{-2/(m+6)})$ for generic neural-network drifts, and a sharper rate $\tilde{O}(n^{-2/(m+4)})$ for one-sided dissipative drift classes, where $n$ is the sample size and $m$ is the dimension of the state space. Numerical experiments provide empirical evidence for the sample-dependent trade-off in $K$, supporting both theoretical regimes. Our framework also suggests a practical implementation principle: choose the diffusion budget $K$ according to the available sample size, and then select a neural-network architecture with the corresponding fixed Lipschitz coefficient.
arXiv:2607.08335v1 Announce Type: cross
Abstract: Policy-based approaches to Bayesian experimental design (BED) allow the learning of deep policy networks that adaptively make intelligent design decisions based on previously collected data. However, the training of such policies is often held back by a fundamental challenge: the double intractability of the expected information gain (EIG). This necessitates expensive or complex approximations that restrict the effort one can invest in optimising the policy itself. To address this, we show that the double intractability of the EIG can be isolated from the policy learning by first solving a score matching problem that is independent of the policy used, then using the learned score approximation to train the policy in a singly intractable manner. This turns the key multiplicative cost into an additive one and reduces the computational burden on the policy training itself, making it far cheaper to train the policy multiple times when needed, e.g. for architecture search, hyperparameter tuning, or avoiding local optima. In our experiments we train multiple competitive policies without inducing a multiplicative cost in likelihood evaluations, which can increase performance by allowing us to select the best policy even without performing hyperparameter or architecture searches.
arXiv:2607.07884v1 Announce Type: new
Abstract: In this short note we consider the gradient descent dynamics of deep scalar linear networks, $f(x) = \prod_{l=1}^L w_l x$, which enjoy exact time-course solutions for any integer depth. We show that even in this minimal model, the optimal depth-wise learning rate scaling depends on data, whereas data-agnostic scaling rules fail to transfer across depths. Under the data-dependent optimal scaling, the learning dynamics is independent of data and weakly dependent on depth, resulting in a constant linear convergence rate across all depths including infinity. We further show similar data-dependent effects in deep scalar linear networks with residual connections.
arXiv:2607.07929v1 Announce Type: new
Abstract: We report precision measurements of the ratio of Land\'e $g$ factors between the $3^2D_{5/2}$ and $4^2S_{1/2}$ states of a single trapped $^{40}\text{Ca}^+$ ion. The measurements are performed in two distinct ion trap apparatus: a cryogenic surface electrode radiofrequency Paul trap and a room-temperature permanent magnet Penning trap. The Penning trap measurements yield a ratio of $0.599~488~813~3(2)$, which represents a more than 40-fold uncertainty reduction compared to previous work. The radiofrequency trap measurement yields a concurring value of $0.599~488~813(6)$. We estimate that systematic shifts for each system are well below the respective statistical uncertainty.
arXiv:2607.07933v1 Announce Type: new
Abstract: Computational modeling plays a central role in advancing our understanding of electro-chemical reactions and may thus guide the rational development of sustainable energy technologies. Despite significant methodological progress, achieving predictive and quantitatively reliable simulations of electrified solid-liquid interfaces under operando conditions remains a major challenge. Recently, the European COST Action EU-CONCERT (EUropean COllaborative Network on electroCatalysis for Efficient Renewable Technologies) was launched with the aim of advancing atomistic modeling of electrochemical reactions through systematic benchmarking and cross-validation. This position paper outlines the consortium's view of the current state of the art, identifies the priority methodological challenges to be addressed, and defines the benchmark tests that are urgently needed. This paper should serve as a guide to the community and a roadmap for the EU-CONCERT COST Action.
arXiv:2607.08351v1 Announce Type: cross
Abstract: Foundation machine-learning interatomic potentials (MLIPs) are rapidly replacing density-functional theory (DFT) for modeling structure and nuclear dynamics, making their fidelity in strongly correlated systems an urgent question. We test three foundation potentials on the low-temperature order of three correlated, isostructural ABO3 perovskite oxides: LaMnO3 (d4), LaCoO3 (d6), and NdNiO3 (d7). We run molecular dynamics for 1 ns on 80- and 160-atom supercells from 50 to 300 K with no system-specific training. These oxides expose three distinct classes of low-temperature order that define a hierarchy of difficulty for the potentials. The scalar class, represented by NdNiO3, has a simple geometric fingerprint and is captured. The vector class, represented by LaMnO3, requires identifying which Cartesian axis carries the long bond at each site, and is captured in magnitude but not in symmetry. The on-site class, represented by the low-spin to high-spin crossover in LaCoO3, is a purely local multiplet population shift with no spatial order parameter and remains inaccessible to present-day MLIPs.
arXiv:2306.12857v3 Announce Type: replace
Abstract: The storage, management, and application of massive spatio-temporal data are widely used in practical scenarios, including public safety. However, due to the unique spatio-temporal distribution characteristics of real-world data, existing methods still face limitations in preserving spatio-temporal proximity and achieving load balancing in distributed storage. This paper proposes an efficient partitioning method for large-scale public safety spatio-temporal data based on information loss constraints, named IFL-LSTP. The model combines a spatio-temporal partitioning module (STPM) and a graph partitioning module (GPM). STPM reduces the scale of data under a predefined information-loss threshold, while GPM uses graph representation learning to obtain balanced graph partitions. Experiments on multiple real-world datasets show that IFL-LSTP can reduce data scale, shorten graph model training time, preserve spatio-temporal proximity, and improve load-balancing effectiveness.
arXiv:2607.08406v1 Announce Type: new
Abstract: Backpropagation (BP) dominates deep learning training, but its reliance on gradients brings inherent troubles -- vanishing and exploding gradients. The pursuit of gradient-free methods has long been a goal in the field of artificial intelligence. This paper shows that indeed the simplest Monte Carlo algorithm implemented on a single GPU -- randomly mutate a parameter, keep it if the loss decreases, otherwise retry -- can practically train deep networks. This gradient-free method does not even need common techniques such as batch normalization or residual connections to directly train sufficiently deep networks. More remarkably, its flexibility extends to several nontrivial scenarios: it enables pure pruning training, supports discrete weights, accommodates unconventional transfer functions such as Gaussian, and reveals the substantial redundancy of deep networks. We have demonstrated its feasibility on deep networks with more than 20 layers, single-hidden-layer wide networks with up to 16,384 hidden neurons, and even a simple Transformer architecture trained on both image classification (MNIST) and character-level language modeling (Tiny Shakespeare). This simple gradient-free method may offer a complementary perspective for understanding the self-organization and learning mechanisms of neural networks, and also provides an alternative route for building physically inspired deep learning systems.
arXiv:2607.08434v1 Announce Type: new
Abstract: Current Unified Large Multimodal Models (ULMMs) support interleaved multimodal reasoning through textual reasoning and intermediate visual states, but typically generate each visual state as a full image. This full-image generation paradigm introduces substantial visual-token redundancy and dilutes supervision on sparse yet reasoning-critical state transitions. We propose DeltaV, a ULMM that replaces full-image generation with visual updates. Conditioned on historical visual states, DeltaV incrementally predicts compact update tokens that capture the visual changes across reasoning steps, avoiding repeated modeling of unchanged content. To align the token budget of each update with the magnitude of visual change, DeltaV introduces a temporal similarity (TSIM) Router, which stops allocating tokens once the marginal reconstruction gain falls below a threshold. To support more diverse and generalizable reasoning, we further construct StructCoT, a large-scale interleaved multimodal reasoning dataset with 1.05M samples spanning 44 task domains. Experiments show that the visual-update paradigm reduces newly generated visual tokens by 55.6\% on average without compromising reconstruction fidelity, and improves multimodal reasoning by 3.3\% over full-image generation. Trained with StructCoT and large-scale multimodal data, DeltaV-2B further outperforms substantially larger open-source models by 8.4\% on in-domain multimodal reasoning evaluations and surpasses the comparable-scale Qwen3-VL-2B by 5.9\% on external multimodal reasoning and understanding benchmarks. Code, models, and StructCoT will be released at https://github.com/Pengjie-W/DeltaV.
arXiv:2310.11375v2 Announce Type: replace
Abstract: Creating learning environments that can accommodate teachers' diverse needs is challenging because responsive elements are not clearly defined or identified. This study identified responsive teacher professional development (PD) elements by taking a phenomenological approach. Using surveys and interviews with 13 high school physics teachers in a PD program at a Midwestern university, we identified responsive features such as practicality, flexibility, and accessibility core to the enactment of a responsive PD. Other features were opportunities for community engagement, pedagogical support, and professional growth, which were aligned with the benefits of engagement in a Community of Practice model incorporated in this work.
arXiv:2511.13999v2 Announce Type: replace
Abstract: We study the running time, in terms of first order oracle queries, of differentially private empirical/population risk minimization of Lipschitz convex losses. We first consider the setting where the loss is non-smooth and the optimizer interacts with a private proxy oracle, which sends only private messages about a minibatch of gradients. In this setting, we show that expected running time $\Omega(\min\{\frac{\sqrt{d}}{\alpha^2}, \frac{d}{\log(1/\alpha)}\})$ is necessary to achieve $\alpha$ excess risk on problems of dimension $d$ when $d \geq 1/\alpha^2$. Upper bounds via DP-SGD show these results are tight when $d>\tilde{\Omega}(1/\alpha^4)$. We further show our lower bound can be strengthened to $\Omega(\min\{\frac{d}{\bar{m}\alpha^2}, \frac{d}{\log(1/\alpha)} \})$ for algorithms which use minibatches of size at most $\bar{m} < \sqrt{d}$. We next consider smooth losses, where we relax the private oracle assumption and give lower bounds under only the condition that the optimizer is private. Here, we lower bound the expected number of first order oracle calls by $\tilde{\Omega}\big(\frac{\sqrt{d}}{\alpha} + \min\{\frac{1}{\alpha^2}, n\}\big)$, where $n$ is the size of the dataset. Modifications to existing algorithms show this bound is nearly tight. Compared to non-private lower bounds, our results show that differentially private optimizers pay a dimension dependent runtime penalty. Finally, as a natural extension of our proof technique, we show lower bounds in the non-smooth setting for optimizers interacting with information limited oracles. Specifically, if the proxy oracle transmits at most $\Gamma$-bits of information about the gradients in the minibatch, then $\Omega\big(\min\{\frac{d}{\alpha^2\Gamma}, \frac{d}{\log(1/\alpha)}\}\big)$ oracle calls are needed. This result shows fundamental limitations of gradient quantization techniques in optimization.
arXiv:2512.15138v2 Announce Type: replace
Abstract: Reference-guided instance editing is fundamentally limited by semantic entanglement, where a reference's intrinsic appearance is intertwined with its extrinsic attributes. The key challenge lies in disentangling what information should be borrowed from the reference, and determining how to apply it appropriately to the target. To tackle this challenge, we propose GENIE, a Generalizable Instance Editing framework capable of achieving explicit disentanglement. GENIE first corrects spatial misalignments with a Spatial Alignment Module (SAM). Then, an Adaptive Residual Scaling Module (ARSM) learns what to borrow by amplifying salient intrinsic cues while suppressing extrinsic attributes, while a Progressive Attention Fusion (PAF) mechanism learns how to render this appearance onto the target, preserving its structure. Extensive experiments on the challenging AnyInsertion dataset demonstrate that GENIE achieves state-of-the-art fidelity and robustness, setting a new standard for disentanglement-based instance editing.
arXiv:2601.02871v3 Announce Type: replace
Abstract: Task-oriented proactive dialogue agents play a pivotal role in recruitment, particularly for steering conversations towards specific business outcomes, such as acquiring social-media contacts for private-channel conversion. Although supervised fine-tuning and reinforcement learning have proven effective for training such agents, their performance is heavily constrained by the scarcity of high-quality, goal-oriented domain-specific training data. To address this challenge, we propose SimRPD, a three-stage framework for training recruitment proactive dialogue agents. First, we develop a high-fidelity user simulator to synthesize large-scale conversational data through multi-turn online dialogue. Then we introduce a multi-dimensional evaluation framework based on Chain-of-Intention (CoI) to comprehensively assess the simulator and effectively select high-quality data, incorporating both global-level and instance-level metrics. Finally, we train the recruitment proactive dialogue agent on the selected dataset. Experiments in a real-world recruitment scenario demonstrate that SimRPD outperforms existing simulator-based data selection strategies, highlighting its practical value for industrial deployment and its potential applicability to other business-oriented dialogue scenarios.
arXiv:2602.12612v2 Announce Type: replace
Abstract: Traditional methods for automating recommender system design, such as Neural Architecture Search (NAS), are often constrained by a fixed search space defined by human priors, limiting innovation to pre-defined operators. While recent LLM-driven code evolution frameworks shift fixed search space target to open-ended program spaces, they primarily rely on scalar metrics (e.g., NDCG, Hit Ratio) that fail to provide qualitative insights into model failures or directional guidance for improvement. To address this, we propose Self-EvolveRec, a novel framework that establishes a directional feedback loop by integrating a User Simulator for qualitative critiques and a Model Diagnosis Tool for quantitative internal verification. Furthermore, we introduce a Diagnosis Tool - Model Co-Evolution strategy to ensure that evaluation criteria dynamically adapt as the recommendation architecture evolves. Extensive experiments demonstrate that Self-EvolveRec significantly outperforms state-of-the-art NAS and LLM-driven code evolution baselines in both recommendation performance and user satisfaction. Our code is available at https://github.com/Sein-Kim/self_evolverec.