Forskningsradar

Science Journals

Peer-reviewade publikationer — 56237 artiklar

arXiv (fysik & CS) Preprint 2026-06-01
Beyond Instance-Level Alignment and Uniformity: Semantic Factor Learning for Collaborative Filtering
arXiv:2605.31414v1 Announce Type: new Abstract: Collaborative filtering (CF) is widely used in recommender systems (RecSys) due to its simplicity and efficiency. However, existing CF methods follow an instance-level learning paradigm. During the instance learning stage, a large number of uninteracted user-item instances, of which items are potential interested by the user, are incorrectly treated as true negative samples resulting in a severe limitation to the generalization and scalability of models. Moreover, mainstream graph convolutional networks (GCNs) inherently suffer from high computational cost and over-smoothing issues, which limit the ability in capturing higher-order connectivity and lead to a poor generalization under sparse supervision signals. To address the above limitations, we propose Semantic Factor enhanced Alignment and Uniformity (SaFeAU), a novel framework that augments interacted instances with semantic factors, thereby mitigating false negative labeling and enabling matrix factorization (MF) to capture high-order CF signals without graph neighborhood aggregation. Specifically, SaFeAU consists of three tightly coupled components. First, Semantic Factor Routing (SFR) disentangles item representations into independent and global semantic factors. Building on these factors, Semantic Factor Matching (SFM) identifies uninteracted items, which share the same semantic factors with interacted ones, as potential positive pairs for enriching sparse supervision signals. Finally, Semantic Pairs Alignment (SPA) aligns both observed and potential positive pairs while promoting uniformity of user and item representations. Extensive experiments on four sparse real-world datasets show that SaFeAU consistently outperforms GCN-based and MF-based state-of-the-art CF methods in both recommendation accuracy and computational efficiency, confirming the effectiveness of the proposed semantic enhanced learning paradigm.
arXiv (fysik & CS) Preprint 2026-06-01
Compute Allocation in Evolutionary Search: From Depth-Breadth to Multi-Armed Bandits
arXiv:2605.29268v2 Announce Type: replace Abstract: LLM-guided evolutionary search (Evolve systems) has reached state-of-the-art results on mathematical and combinatorial tasks, yet most existing systems report only the best of many runs and leave the run-to-run distribution undocumented. We ask how a fixed budget of LLM calls should be allocated, and how reliably a single run reaches the reported numbers. Sweeping the depth-breadth grid over five models and three tasks, we identify two empirical regularities: a fitness-compute envelope along which capability ordering largely collapses on effective FLOPs, and a bilinear depth-breadth fit with task-specific interaction; both are gated by model-task capability. Motivated by these regularities, we propose BaSE (Bandit-based Self-Evolving), a multi-armed bandit that allocates LLM calls across parallel trajectories. Without changing the model, prompt, or evaluator, BaSE improves mean fitness by 12.3% over the strongest island-protocol baseline across 8 (model, task) cells, with the largest gains on high-variance settings: a reliability gain from allocation alone.
arXiv (fysik & CS) Preprint 2026-06-01
Domain-Specific Data Synthesis for LLMs via Minimal Sufficient Representation Learning
arXiv:2605.30039v2 Announce Type: replace Abstract: Large Language Models have demonstrated remarkable progress in general-purpose capabilities and can achieve strong performance in specific domains through fine-tuning on domain-specific data. However, acquiring high-quality data for target domains remains a significant challenge. Existing data synthesis approaches follow a deductive paradigm, heavily relying on explicit domain descriptions expressed in natural language and careful prompt engineering, limiting their applicability in real-world scenarios where domains are difficult to describe or formally articulate. In this work, we tackle the underexplored problem of domain-specific data synthesis through an inductive paradigm, where the target domain is defined only through a set of reference examples, particularly when domain characteristics are difficult to articulate in natural language. We propose a novel framework, DOMINO, that learns a minimal sufficient domain representation from reference samples and leverages it to guide the generation of domain-aligned synthetic data. DOMINO integrates prompt tuning with a contrastive disentanglement objective to separate domain-level patterns from sample-specific noise, mitigating overfitting while preserving core domain characteristics. Theoretically, we prove that DOMINO expands the support of the synthetic data distribution, ensuring greater diversity. Empirically, on challenging coding benchmarks where domain definitions are implicit, fine-tuning on data synthesized by DOMINO improves Pass@1 accuracy by up to 4.63\% over strong, instruction-tuned backbones, demonstrating its effectiveness and robustness. This work establishes a new paradigm for domain-specific data synthesis, enabling practical and scalable domain adaptation without manual prompt design or natural language domain specifications.
arXiv (fysik & CS) Preprint 2026-06-01
FAM-Bench: A Multimodal Benchmark for Condition-Aware Food-as-Medicine Reasoning
arXiv:2605.31410v1 Announce Type: new Abstract: Food-as-Medicine requires models to reason beyond what a dish is or what nutrition it contains: they must decide whether a concrete food choice is appropriate for a specific health condition. Existing food AI benchmarks primarily evaluate dish recognition, recipe understanding, nutrient estimation, or general nutrition question answering, leaving this health-aware decision layer largely untested. We introduce FAM-Bench, a multi-modal Food-as-Medicine benchmark with 2500 nutrition-expert-verified instances across 13 diet-related health conditions. The benchmark contains two complementary tasks: dish-level suitability assessment, where models judge whether a dish is suitable for a condition from its image and ingredient list, and comparative dish analysis, where models rank four candidate dishes by condition-specific suitability. Both tasks require integrating ingredient evidence, visual preparation cues, and clinical nutrition constraints, providing a standardized testbed for grounded health-aware reasoning in language and vision-language models.
arXiv (fysik & CS) Preprint 2026-06-01
Surface Constraint Policy for Learning Surface-Constrained and Dynamically Feasible Robot Skills
arXiv:2605.31321v1 Announce Type: new Abstract: Diffusion-based imitation learning methods have driven rapid progress in robot dexterous manipulation tasks. However, they have limitations when applied to tasks that involve complex free-form surface constraints because of their lack of explicit surface geometry constraint modeling and the dynamic feasibility issue, resulting in stochastic action generation that fails to achieve reliable surface alignment and maintain stable contact. To address these limitations, we propose a novel surface constraint policy (SCP) for generating robot actions that satisfy free-form surface constraints on the basis of human demonstrations and real-time visual observations. First, the surface geometry constraint is encoded using a two-dimensional weighted Gaussian kernel function that is derived from demonstrations. Building on the encoded surface geometry constraints, the diffusion-based policy is used to infer task-level action intentions from multimodal sensory inputs, including visual observations and robot state feedback. These intentions are further transformed into surface-constrained dynamic movement primitives (DMPs) through a similarity-based action mapping method, thereby enabling smooth and compliant motion execution. The SCP achieves generation of structured surface geometric intent and dynamically admissible actions. The proposed method is validated on multiple surface manipulation tasks and compared with existing techniques. The experimental results demonstrate superior task success rates and contact stability under surface constraints.
arXiv (fysik & CS) Preprint 2026-06-01
Governance-Aware Software Architecture for Multi-Stakeholder Platforms
arXiv:2605.31316v1 Announce Type: new Abstract: Multi-stakeholder platforms (MSPs) coordinate diverse stakeholder groups, often with competing or conflicting requirements. As these platforms increasingly take digital form, engineers building them make architectural decisions about data visibility, service decomposition, and algorithm design that directly determine which stakeholder requirements are prioritized when conflicts arise. Software architecture literature provides patterns for data isolation and access control among tenants but does not address how architectural decisions resolve conflicts among stakeholders with structurally divergent interests. MSP governance literature identifies the principles at stake but treats technology as neutral infrastructure. Neither addresses the translation between governance principles and architectural decision spaces. This paper proposes a governance-architecture correspondence framework that surfaces implicit governance decisions, making them explicit and debatable before deployment. The framework maps five MSP governance principles to the architectural decision spaces where they must be addressed, identifying for each the governance-aware design choice and the technically convenient default it overrides. We illustrate the framework in a constructed knowledge platform for pig farming in Rwanda, where five stakeholder types present structurally conflicting requirements. As work in progress, the framework is proposed but not yet empirically validated; a planned pre/post judgment study with platform users across all stakeholder types will test falsifiable predictions about governance outcomes.
arXiv (fysik & CS) Preprint 2026-06-01
Learning Parametric Nitrogen Fertilizer Response Curves Using Neuro Symbolic Regression
arXiv:2605.31276v1 Announce Type: new Abstract: Accurately modeling crop response to Nitrogen (N) fertilization is a fundamental challenge in precision agriculture, as it impacts both economic returns and environmental sustainability. Existing approaches either rely on predefined parametric forms or opaque machine learning models, limiting their ability to interpret or discover site-specific functional relationships from data. In this work, we propose a neuro symbolic regression (SR) approach to learn parametric N-response curves without assuming a predefined functional form. Our approach integrates a transformer-based Multi-Set Symbolic Skeleton Prediction strategy, enabling the discovery of shared functional structures across multiple subdomains or management zones (MZs). By constructing diverse input subsets and enforcing consistency across them, the method recovers robust symbolic skeletons that are subsequently fitted to observed data using a genetic algorithm. This framework was first evaluated on synthetic one-dimensional problems to assess its robustness under varying levels of epistemic uncertainty. The results demonstrate the ability of the proposed SR approach to recover correct expressions even in data-scarce regimes. In this work, we present the results of applying our method to real-world winter wheat data, learning distinct parametric N-response curves for different MZs within a field. The results show that the discovered expressions not only achieve lower fitting errors than traditional models such as quadratic-plateau and exponential functions, but also capture diverse functional behaviors across spatial regions. This demonstrates the potential that neuro SR has to enable the discovery of site-specific agronomic relationships and support informed decision-making in precision agriculture.
arXiv (fysik & CS) Preprint 2026-06-01
Re-examining Granger Causality with Causal Bayesian Networks and Reichenbachs Principles
arXiv:2501.02672v3 Announce Type: replace-cross Abstract: Characterising cause-effect relationships in complex systems is fundamental to understanding their underlying mechanisms. Granger causality (GC) remains a widely used computational tool for identifying causal relationships in time series data. However, like other causal discovery methods, GC has limitations and has been criticised for lacking a rigorous causal foundation. In this work, we present a fix to this criticism by reinterpreting GC through the lenses of Reichenbach's principles and causal Bayesian networks. This reinterpretation was implemented as an algorithm we call causalized Granger causality (c-GC). We demonstrate, both theoretically and graphically, that this reformulation endows GC with a robust causal interpretation under specific assumptions. c-GC yields satisfactory results on synthetic data, offering a more principled framework for causal discovery in observational datasets.
arXiv (fysik & CS) Preprint 2026-06-01
Steering Fractional-Order Network Dynamics via Joint Parameter and State Control
arXiv:2605.31270v1 Announce Type: new Abstract: This paper studies the control of discrete-time linear fractional-order networks, a flexible modeling framework for systems with long-range memory such as power grids, biological networks, and neuronal circuits. In contrast to the common view that fractional exponents (time-scales) are fixed parameters, we show that they can be systematically steered, together with the network coupling matrix, by appropriately designed input sequences. We first derive algebraic conditions under which the coupling matrix and the vector of fractional exponents of a given network can be reconfigured to desired values, and we characterize how truncating the infinite-memory term impacts the resulting dynamics. Building on these results, we construct an equivalent linear representation that isolates the contribution of memory, and we introduce a fractional reachability matrix that provides explicit conditions for jointly steering both network parameters and state in a finite number of steps. To address practical implementations, we further formulate an energy-constrained steering problem that incorporates actuator bounds and finite-memory approximations as a quadratic program. The framework is illustrated on low-dimensional toy examples, on larger networks with Erdos-Renyi, Barabasi-Albert, and Watts-Strogatz topologies, and on a brain network model inferred from electrocorticography recordings of an epilepsy patient, where we showcase transitions between pre-seizure and seizure configurations.
arXiv (fysik & CS) Preprint 2026-06-01
Safe Arrival Scheduling at Constraint Waypoints in UAM Corridors
arXiv:2605.31243v1 Announce Type: new Abstract: This study introduces a novel Air Traffic Control (ATC) concept to support self-separation between vehicles in Urban Air Mobility (UAM) corridors. Our proposed scheme involves sharing intended arrival schedules at Constrained Waypoints (CWPs) among UAM operators. We propose two approaches to assist the arrival scheduling at CWPs by computing the minimum arrival time gap necessary for each pair of vehicles to ensure their safety throughout the flights within the corridor. The first approach considers the minimum separation distance required by the Near Mid-Air-Collision (NMAC) avoidance rules, while the second one is based on the Responsibility-Sensitive Safety (RSS) rules. We demonstrate that the NMAC-rule-based approach can effectively prevent collisions in normal circumstances, where the vehicles adhere to the speed limits of the corridor. However, this approach does not guarantee safety if vehicles exceed the speed limits. Conversely, while the RSS-rule-based approach ensures collision prevention during emergencies when vehicles exceed speed limits, it may require larger arrival time gaps under normal circumstances, which may lead to reduced traffic flow. Our results are confirmed through numerical simulations.
arXiv (fysik & CS) Preprint 2026-06-01
Adaptive Physics Transformer with Fused Global-Local Attention for Subsurface Energy Systems
arXiv:2602.11208v2 Announce Type: replace Abstract: The Earth's subsurface is a cornerstone of modern society, providing essential energy resources like hydrocarbons, geothermal, and minerals while serving as the primary reservoir for $CO_2$ sequestration. However, full physics numerical simulations of these systems are notoriously computationally expensive due to geological heterogeneity, high resolution requirements, and the tight coupling of physical processes with distinct propagation time scales. Here we propose the $\textbf{Adaptive Physics Transformer}$ (APT), a geometry-, mesh-, and physics-agnostic neural operator that explicitly addresses these challenges. APT fuses a graph-based encoder to extract high-resolution local heterogeneous features with a global attention mechanism to resolve long-range physical impacts. Our results demonstrate that APT outperforms state-of-the-art architectures in subsurface tasks across both regular and irregular grids with robust super-resolution capabilities. Notably, APT is the first architecture that learns directly from HR-adaptive mesh refinement simulations. We also demonstrate APT's favorable scaling behavior and cross-dataset learning capability, positioning it as a robust and scalable backbone for large-scale subsurface foundation model development.
arXiv (fysik & CS) Preprint 2026-06-01
LaCy: What Small Language Models Can and Should Learn is Not Just a Question of Loss
arXiv:2602.12005v4 Announce Type: replace Abstract: Language models have consistently grown to compress more world knowledge into their parameters, but the knowledge that can be pretrained into them is upper-bounded by their parameter size. Especially the capacity of Small Language Models (SLMs) is limited, leading to factually incorrect generations. This problem is often mitigated by giving the SLM access to an outside source: the ability to query a larger model, documents, or a database. Under this setting, we study the fundamental question of \emph{which tokens an SLM can and should learn} during pretraining, versus \emph{which ones it should delegate} via a \texttt{<CALL>} token. We find that this is not simply a question of loss: although the loss is predictive of whether a predicted token mismatches the ground-truth, it is insufficient for identifying which predictions would actually lead to factual or semantically invalid continuations. Some high-loss tokens correspond to \emph{acceptable} alternative continuations of a pretraining document and therefore should not trigger a \texttt{<CALL>}. This suggests that learnability cannot be characterized from loss alone, but requires additional domain-specific signals about the role of a token in the sentence. In Wikipedia-like domains, we show that augmenting the loss signal with lightweight grammatical information from a spaCy parser substantially improves delegation decisions. Based on this insight, we propose LaCy, a novel pretraining method that combines loss with factuality signals to decide which tokens an SLM should learn. Our experiments demonstrate that LaCy models successfully learn which tokens to predict and when to call for help. This results in higher FactScores when generating in a cascade with a bigger model and outperforms Rho or LLM-judge trained SLMs, while being simpler and cheaper.
arXiv (fysik & CS) Preprint 2026-06-01
The Information Geometry of Softmax: Probing and Steering
arXiv:2602.15293v2 Announce Type: replace Abstract: This paper concerns the question of how AI systems encode semantic structure into the geometric structure of their representation spaces. The motivating observation is that the natural geometry of these representation spaces should reflect the way models use representations to produce behavior. We focus on the important special case of representations that define softmax distributions. In this case, we argue that the natural geometry is information geometry. Our focus is on the role of information geometry on semantic encoding and the linear representation hypothesis. As an illustrative application, we develop "dual steering", a method for robustly steering representations to exhibit a particular concept using linear probes. We prove that dual steering optimally modifies the target concept while minimizing changes to off-target concepts. Empirically, we find that dual steering enhances the controllability and stability of concept manipulation.
arXiv (fysik & CS) Preprint 2026-06-01
Beyond ReLU: Bifurcation, Oversmoothing, and Topological Priors
arXiv:2602.15634v2 Announce Type: replace Abstract: Graph Neural Networks (GNNs) learn node representations through iterative network-based message-passing. While powerful, deep GNNs suffer from oversmoothing, where node features converge to a homogeneous, non-informative state. We re-frame this problem of representational collapse from a \emph{bifurcation theory} perspective, characterizing oversmoothing as convergence to a stable ``homogeneous fixed point.'' Our central contribution is the theoretical discovery that this undesired stability can be broken by replacing standard monotone activations (e.g., ReLU) with a class of functions. Using Lyapunov-Schmidt reduction, we analytically prove that this substitution induces a bifurcation that destabilizes the homogeneous state and creates a new pair of stable, non-homogeneous \emph{patterns} that provably resist oversmoothing. Our theory predicts a precise, nontrivial scaling law for the amplitude of these emergent patterns, which we quantitatively validate in experiments. Finally, we demonstrate the practical utility of our theory by deriving a closed-form, bifurcation-aware initialization and showing its utility in real benchmark experiments.
arXiv (fysik & CS) Preprint 2026-06-01
Efficient Benchmarking Is Just Feature Selection and Multiple Regression
arXiv:2605.25773v2 Announce Type: replace-cross Abstract: Efficient benchmarking techniques aim to lower the computational cost of evaluating LLMs by predicting full benchmark scores using only a subset of a benchmark's questions. By reframing this problem as an instance of multiple regression with feature selection, we find that existing efficient benchmarking methods can be greatly improved by simply using kernel ridge regression at the prediction stage. Additionally, using an information-theoretic feature-selection algorithm called minimum redundancy maximum relevance (mRMR), we can further improve upon these methods by selecting question subsets that will be maximally useful for prediction. Except in very data-poor settings, these approaches consistently achieve smaller prediction errors (in both MAE and RMSE), and greater ranking correlation between predicted and true scores (in both Spearman $\rho$ and Kendall $\tau$) across a range of benchmarks using both binary and continuous metrics. Furthermore, mRMR subsampling is much faster than competitor methods (which often involve fitting probabilistic models or running clustering algorithms), and is more likely to select the same questions under different random seeds or training data splits. Tutorial code can be found at https://github.com/sambowyer/mrmr_eval .
arXiv (fysik & CS) Preprint 2026-06-01
Feedback Control of a Recirculating Bioreactor with Electrophoretic Removal of Inhibitory Extracellular DNA
arXiv:2603.23150v2 Announce Type: replace Abstract: Extracellular DNA accumulation in recirculating bioprocesses inhibits microbial growth and reduces productivity. We consider a continuous bioreactor with a recirculating loop and an electrophoretic filtration unit for selective DNA removal, and develop a feedback control framework combining online state and parameter estimation via an Unscented Kalman Filter with an adaptive Model Predictive Controller that jointly optimizes dilution rate and filtration activation. Closed-loop simulations under nominal and perturbed conditions show that the addition of the filtration unit enables the proposed control strategy to achieve significantly higher cumulative profit while keeping DNA concentration below the inhibition threshold.
arXiv (fysik & CS) Preprint 2026-06-01
Improved Approximation Algorithms and Hardness Results for Shortest Common Superstring with Reverse Complements
arXiv:2603.26176v2 Announce Type: replace Abstract: The Shortest Common Superstring (SCS) problem is a fundamental task in sequence analysis. In genome assembly, however, the double-stranded nature of DNA implies that each fragment may occur either in its original orientation or as its reverse complement. This motivates the Shortest Common Superstring with Reverse Complements (SCS-RC) problem, which asks for a shortest string that contains, for each input string, either the string itself or its reverse complement as a substring. The previously best-known approximation ratio for SCS-RC was $\frac{23}{8}$. In this paper, we present a new approximation algorithm achieving an improved ratio of $\frac{8}{3}$. Our approach computes an optimal constrained cycle cover by reducing the problem, via a novel gadget construction, to a maximum-weight perfect matching in a general graph. We also investigate the computational hardness of SCS-RC. While the decision version is known to be NP-complete, no explicit inapproximability results were previously established. We show that the hardness of SCS carries over to SCS-RC through a polynomial-time reduction, implying that it is NP-hard to approximate SCS-RC within a factor better than $\frac{333}{332}$. Notably, this hardness result holds even for the DNA alphabet.
arXiv (fysik & CS) Preprint 2026-06-01
Serverless5GC: Private 5G Core Deployment via a Procedure-as-a-Function Architecture
arXiv:2603.27618v2 Announce Type: replace Abstract: Open-source 5G core implementations deploy network functions as always-on processes that consume resources even when idle. This inefficiency is most problematic in private and edge deployments with sporadic traffic. Serverless5GC is an architecture that maps each 3GPP control-plane procedure to an independent Function-as-a-Service invocation, allowing scale-to-zero operation without modifying the standard N2 interface. The architecture decomposes 12 network functions (3GPP Release 15-17) into 31 serverless procedures, fronted by an SCTP/NGAP proxy that bridges unmodified RAN equipment to an HTTP-based serverless backend. Evaluation against Open5GS and free5GC across five traffic scenarios (idle to 50 registrations/s burst) shows that Serverless5GC achieves median registration latency of 406-522 ms, on par with the C-based Open5GS baseline (403-606 ms), while maintaining 100% success across 3,000 registrations. A resource-time cost model shows that the serverless deployment (0.002 GB-seconds per registration) is cheaper than the always-on baseline when the cluster operates below a 0.65 duty cycle, when two or more tenants share the platform, or on managed FaaS platforms up to 595 reg/s. Under worst-case cold-start conditions where all 31 function pods are evicted simultaneously, the system sustains zero failures and converges to warm-start latency within 4-5 seconds.
arXiv (fysik & CS) Preprint 2026-06-01
Effects of preferential concentration on the combustion of iron particles -- A numerical study with homogeneous isotropic turbulence
arXiv:2604.04048v2 Announce Type: replace Abstract: Iron particles, with their non-volatile combustion mode, remain in the dispersed phase throughout the combustion process, causing the flow in a typical iron powder combustor to be particle-laden and turbulent. Preferential concentration is a phenomenon prevalent in such turbulent flows that causes particle clustering. To estimate the effects of clustering on the combustion process, direct-numerical-simulations are performed on a cubical domain with forced homogeneous isotropic turbulence. Simulations pertaining to Kolmogorov Stokes number $\mathrm{St}=1,10,50$, turbulent Reynolds number $\mathrm{Re_\lambda}= 5,10,20$, and global equivalence ratio (considering FeO as the oxidation product) $\phi=0.25,0.5,0.75$ are executed. Increasing $\phi$ significantly extends the combustion completion time. A Poisson distribution of particles burns faster with a higher peak mean temperature. The evolution of the mean temperature in the combustion of the clustered distribution is smooth and results in a smaller peak value. However, the total combustion time of a clustered distribution is significantly extended, by up to eight times at $\mathrm{Re_\lambda}=20$ and $\phi=0.75$. Analysis of the Voronoi volumes $V_\mathrm{norm}$ at the start of combustion shows that particles in highly dense regions burn longer, as seen before in the literature. Furthermore, the combustion time exhibits a strong exponential dependence on $V_\mathrm{norm}$ in the ``cluster'' regions, and an asymptotic behavior in the ``void'' regions. However, significant spread is observed in the correlation. Time-averaging $V_\mathrm{norm}$ does not minimize this variation considerably. Analysis of the macroscale $\mathrm{O_2}$ depletion zone indicates the importance of the macrostructure -- proximity of multiple clusters -- on the extension of the combustion time.
arXiv (fysik & CS) Preprint 2026-06-01
Posterior and Likelihood Sensitivity in Bayesian Distributionally Robust Optimization
arXiv:2605.31306v1 Announce Type: cross Abstract: We introduce the notion of worst-case posterior and worst-case likelihood sensitivity. These measure, respectively, the sensitivity of the expected cost to worst-case perturbations of the posterior distribution and worst-case perturbations of the likelihood of a Bayesian model. Each defines a quantitative measure of robustness. A decision maker concerned about the sensitivity of the out-of-sample expected cost to deviations from her assumptions will want a decision for which both sensitivities are small. We derive posterior and likelihood sensitivities for uncertainty sets defined in terms of deviation measures. Posterior sensitivity vanishes when the posterior variance shrinks to zero, which occurs when parameter uncertainty is eliminated from learning. Parameter learning does not eliminate likelihood sensitivity. A distributionally robust formulation of a Bayesian optimization problem makes a near-Pareto-optimal tradeoff between performance (expected cost) and robustness (posterior and likelihood sensitivity).
arXiv (fysik & CS) Preprint 2026-06-01
Hyper-DP3: Frequency-Aware Right-Sizing of 3D Diffusion Policies for Visuomotor Control
arXiv:2605.01581v4 Announce Type: replace Abstract: Diffusion-based visuomotor policies perform well in robotic manipulation, yet current methods still inherit image-generation-style decoders and multi-step sampling. We revisit this design from a frequency-domain perspective. Robot action trajectories are highly smooth, with most energy concentrated in a few low-frequency discrete cosine transform modes. Under this structure, we show that the error of the optimal denoiser is bounded by the low-frequency subspace dimension and residual high-frequency energy, implying that denoising error saturates after very few reverse steps. This also suggests that action denoising requires a much simpler denoising model than image generation. Motivated by this insight, we propose Hyper-DP3 (HDP3), a pocket-scale 3D diffusion policy with a lightweight Diffusion Mixer decoder that supports two-step DDIM inference. Our synthetic experiments validate the theory and support the sufficiency of two-step denoising. Futhermore, across RoboTwin2.0, Adroit, MetaWorld, and real-world tasks, HDP3 achieves state-of-the-art performance with fewer than 1% of the parameters of prior 3D diffusion-based policies and substantially lower inference latency.
arXiv (fysik & CS) Preprint 2026-06-01
Not All Synthetic Data Is Yours to Learn From
arXiv:2605.31126v1 Announce Type: new Abstract: Can a language model improve from plain text sampled from itself, with no prompts, no teacher, no verifier, and no reward model? Yes, but only when the synthetic corpus is compatible with the student, a relational property of the source-student pair rather than an intrinsic property of the data. We call this the latent capability resurfacing hypothesis: weak self-training can amplify capabilities already present in the pretrained model, but only under this compatibility condition. We study this in the minimal setting of prompt-free unconditional self-training, where base language models are fine-tuned on text generated from the BOS token alone, with no task specification or external supervision. We report three findings. First, synthetic utility is relational rather than intrinsic: self-generated data is the most effective source, same-lineage transfer outperforms stronger but differently trained sources, and cross-family transfer is substantially weaker. Second, common intrinsic proxies fail: neither benchmark-level semantic similarity nor average per-token likelihood under the student predicts which corpora help. Third, this regime produces a surprising byproduct. In controlled Pythia experiments, capability and verbatim memorization decouple: benchmark utility is preserved or improved while held-out exact-match extraction drops by over 95 percent, with no forget set, privacy objective, or targeted unlearning. Together, these results suggest that prompt-free self-training works by amplifying what the student already knows, not by importing structure from the data. They also reveal a regime in which capability and verbatim memorization can be separated without any explicit unlearning objective.
arXiv (fysik & CS) Preprint 2026-06-01
FedQueue: Queue-Aware Federated Learning for Cross-Facility HPC Training
arXiv:2605.02125v3 Announce Type: replace Abstract: Federated learning (FL) across multiple HPC facilities faces stochastic admission delays from batch schedulers that dominate wall-clock time. Synchronous FL suffers from severe stragglers, while asynchronous FL accumulates stale updates when queues spike. We propose FedQueue, a queue-aware FL protocol that incorporates scheduler delays directly into training and aggregation, which (i) predicts per-facility queue delays online to budget local work, (ii) applies cutoff-based admission that buffers late arrivals to bound staleness, and (iii) performs staleness-aware aggregation to stabilize heterogeneous local workloads. We prove the convergence for non-convex objectives at rate $\mathcal{O}(1/\sqrt{R})$ under bounded staleness, and show that the admission controls yield bounded staleness with high probability under queue-prediction error. Real-world cross-facility deployment of FedQueue shows 20.5% improvement over baseline algorithms. Controlled queue simulations demonstrate robust improvement over the baselines; in particular, up to 60% reduction in time to reach a target accuracy level under high queue variance and non-IID partitions.
arXiv (fysik & CS) Preprint 2026-06-01
Tensor gradient flow for rod-like liquid crystals from molecular model with closure approximation by quasi-entropy
arXiv:2605.30735v1 Announce Type: cross Abstract: In tensor dynamics for liquid crystals derived from molecular models, a common problem is closure approximation. For rod-like molecules, the Bingham closure has proved to outperform other methods because it inherits the gradient flow structure of the molecular model, but is difficult to achieve efficient computations maintaining the gradient flow structure. We propose a closure approximation by the quasi-entropy that has been successfully applied to the free energy, based on which we construct the tensor gradient flow. The quasi-entropy closure has the same symmetry properties as the Bingham closure. The resulting tensor gradient flow is able to constrain the eigenvalues of the tensor within the physical range, guaranteeing the positive definiteness of the dissipation operator given by the higher-order tensors. The quasi-entropy closure is easy to implement since it can be reduced to minimizing an elementary function of three variables. As a result, we construct a numerical scheme preserving the eigenvalue constraints and energy dissipation, with the closure approximation decoupled from solving the scheme. Numerical simulations are carried out for the interface between the isotropic and the uniaxial nematic phase, as well as the defect evolutions, where the higher-order tensors indeed make a difference.
arXiv (fysik & CS) Preprint 2026-06-01
Improved Distribution Estimation in $\ell_\infty$
arXiv:2605.30509v1 Announce Type: cross Abstract: We present improved bounds for estimating discrete probability distributions under the $\ell_\infty$ norm. These include minimax bounds in expectation and high-probability tail bounds. We resolve some of the open questions posed in Kontorovich and Painsky (JMLR, 2025) -- including a fully empirical version of the tightest risk bound they presented and identifying the form of the worst-case extremal distribution. Encouraging empirical results are reported as well.