arXiv:2604.18701v3 Announce Type: replace Abstract: Local prediction-error-based curiosity rewards focus on the current transition without considering the world model's cumulative prediction error across all visited transitions. We introduce Curiosity-Critic, which grounds its intrinsic reward in the improvement of this cumulative objective, and show that it admits a tractable per-step surrogate: the difference between the current prediction error and the asymptotic error baseline of the current state transition. We estimate this error baseline online with a learned critic co-trained alongside the world model; since the critic only has to learn how hard a transition is to predict, its estimate of the irreducible noise floor converges well before the world model saturates, redirecting exploration toward learnable transitions. The reward is higher for learnable transitions and collapses toward zero for stochastic ones, thereby separating epistemic (reducible) from aleatoric (irreducible) prediction error online. Prior prediction-error curiosity formulations, from Schmidhuber (1991) to learned-feature-space variants, emerge as special cases corresponding to specific approximations of this error baseline. Experiments on a stochastic grid world show that Curiosity-Critic outperforms prediction-error, visitation-count, and Random Network Distillation methods in training speed and final world model accuracy.
Science Journals
arXiv:2604.22146v3 Announce Type: replace Abstract: Coflow has emerged as a fundamental application-layer abstraction in distributed systems, enabling collaborative management of related flows to enhance job completion efficiency. To meet the increasing bandwidth demands of modern data center networks (DCNs), optical circuit switches are widely deployed due to their high capacity and energy efficiency. Simultaneously, DCN deployments are evolving towards heterogeneous parallel architectures, where multiple independent optical circuit switching (OCS) cores operate concurrently to facilitate bandwidth expansion and incremental upgrades. However, existing research on coflow scheduling in multi-core switching fabrics primarily focuses on electrical packet switching (EPS) networks, with a few known results on OCS networks without or with a poor performance guarantee. This paper studies the coflow scheduling problem in multi-core OCS networks under the not-all-stop reconfiguration model, focusing on two major challenges of overcoming cross-core coupling for inter-core traffic allocation and satisfying the constraints of port exclusivity and reconfiguration overhead for intra-core circuit scheduling. To minimize total weighted coflow completion time (CCT), we propose an efficient algorithm by integrating LP-guided global coflow ordering, inter-core flow allocation and intra-core circuit scheduling that achieves approximation ratios of $8K$ and $\left(8K+1\right)$ for zero and arbitrary release times of coflows, respectively, where $K$ is the number of OCS cores. This framework is also applicable to $H$-core EPS networks, providing approximation guarantees of $4H$ and $\left(4H+1\right)$ for zero-time and arbitrary-time release, respectively.
arXiv:2606.18117v1 Announce Type: new Abstract: The problem of k-way hypergraph partitioning is fundamental with significant applications in various fields, including VLSI design and scientific computing. State-of-the-art hypergraph partitioners commonly employ a multi-level framework encompassing coarsening, initial partitioning, uncoarsening, and refinement phases. However, many existing methods do not scale well to problems requiring a large number of partitions (i.e., large k). In pursuit of exceptionally high solution quality, existing memetic approaches often execute their two key operations, recombination and mutation, by invoking separate, standalone multi-level partitioners. This design choice, however, renders them significantly more time-consuming than standard multi-level partitioners. To make such memetic approaches more practical, we propose an advanced memetic framework, IMPart, which introduces novel recombination and mutation operators and integrates them directly into the uncoarsening phase of a single multi-level framework. This transforms the local searches of different granularities in the traditional multi-level framework into a sophisticated, collaborative search. Experimental results on multiple standard benchmarks demonstrate our framework more effectively escapes local optima and explores the global solution space for higher-quality solutions, substantially outperforming all existing hypergraph partitioners for large-$k$-way hypergraph partitioning. Our framework highlights a new paradigm for the development of advanced hypergraph partitioners.
arXiv:2606.18119v1 Announce Type: new Abstract: To assess the ability of current AI systems to correctly solve research-level mathematics problems, we tested several AI systems on a set of ten problems in a broad range of mathematical fields; these problems arose naturally in the research process of the contributors. This document includes the problems, our methodology, and the results of our testing. We provide links to supplementary documents including the human solutions, the AI-generated solutions, and the referee reports and logs for the AI-generated solutions. The ten problems were contributed by the following mathematicians: (1) Dariusz Kaloci\'nski and Theodore A. Slaman, (2) Richard Schwartz, (3) Aleksa Milojevic and Benny Sudakov, (4) Larry Guth, (5) Oleg Butkovsky, Jonathan Mattingly, and Lorenzo Zambotti, (6) Joshua Evan Greene and Duncan McCoy, (7) Sucharit Sarkar, (8) Sam Payne and Jidong (Jayden) Wang, (9) Sylvie Corteel and John Lentfer, (10) Srivatsav Kunnawalkam Elayavalli.
arXiv:2606.15617v2 Announce Type: replace Abstract: Interpretability is essential for trustworthy medical image diagnosis. However, existing concept-driven interpretable methods have key limitations: Concept Bottleneck Models (CBMs) require scoring all predefined concepts at inference time and for manual intervention, imposing a substantial burden on clinicians, while rationale-based generative approaches often select concepts by class discriminability, which can drift from diagnostic ontologies. To address these issues, we propose Neuro-Symbolic Rule Distillation (NeRD), a framework that produces efficient, ontology-grounded reasoning chains that are sufficient yet non-redundant, without manually crafting diagnostic rules. Experiments on two skin datasets demonstrate strong diagnostic performance and interpretability, and blinded expert evaluation confirms the clinical plausibility of NeRD rationales. Our method further enables a first expert-in-the-loop study for Multimodal Chain-of-Thought-based diagnosis, achieving efficient and effective concept-level intervention.
arXiv:2606.18147v1 Announce Type: new Abstract: Language models are remarkably capable at medical question answering, in some cases surpassing the accuracy of general physicians. However, answering questions about wearable health data remains challenging and understudied, as these ubiquitous sensors produce continuous, high-dimensional, and longitudinal data, which is non-trivial to align with text-centric distributions in LLM pretraining. The diversity of sensor modalities and user intents cannot be effectively handled by a fixed reasoning workflow or a single pretrained foundation model. To address these challenges, we propose WEQA, a query-adaptive agent framework that unifies LLM reasoning with specialized wearable analytical and modeling tools. An LLM controller is employed to synthesize execution plans and dynamically route each query to the appropriate combination of sensor analysis and pretrained models, and perform grounded response auditing with external knowledge. We also curate a benchmark spanning four open wearable datasets comprising analytic and predictive tasks in three different health domains. Experiments show that our framework is 24% more accurate than LLM and agentic baselines, and a blinded study with 12 medical experts and 8 users shows substantial gains in usefulness and clinical soundness.
arXiv:2507.09150v2 Announce Type: replace Abstract: High frequency electrostatic drift waves are shown as hybrid modes generated by mixing with cyclotron dynamics and grow at the expense of cyclotron modes due to mode coupling effects. The role of density gradient is to break the symmetry between the two cyclotron branches with strong values of the gradient causing the branch modified by drift to overlap with the cyclotron branch leading to a mode coupling instability. The analysis is carried out in the framework oof a fluid model without invoking collisional or finite Larmor radius effects. A second order nonlinear equation with variable coefficients has been derived to govern the dynamics of high frequency electrostatic drift waves in a moving frame of (2+1) spatio-temporal dimensions to show the wave characteristics in various parameter ranges. The possibilities of excitation of the high frequency electrostatic drift waves have been explored in the context of certain laboratory and astrophysical plasma systems including tokamaks and solar corona where the existence of steeper density gradients is more probable.
arXiv:2605.31286v2 Announce Type: replace Abstract: Real-world household robots require Vision-Language-Action (VLA) foundation models that can acquire reusable manipulation skills across diverse objects, task conditions, and household environments. Deformable-object folding is a representative challenge, requiring robots to handle clothing items from random initial states across varying categories, geometries, materials, and scenes. However, existing VLA systems commonly train separate policies for different object categories, while naively mixed multi-task training often suffers from task interference and degraded performance. To move beyond category-specific folding policies, we introduce DeMaVLA, a VLA foundation model for generalizable Deformable Manipulation. DeMaVLA adopts a VLM backbone with an action expert and formulates continuous action generation using flow matching. To improve efficiency, the action expert is constructed by pruning every other transformer layer while preserving layer-wise alignment with the VLM backbone, reducing training and inference cost. DeMaVLA is first pre-trained on approximately 5,000 hours of selected real-world dual-arm demonstrations to acquire general manipulation priors. It is then post-trained on mixed folding data that aggregates self-collected demonstrations and corrective trajectories from real-robot failures across multiple folding tasks through a human-in-the-loop Data Aggregation~(DAgger) pipeline. Experiments show that DeMaVLA achieves competitive performance on RoboTwin 2.0 and strong real-world results on our household folding benchmark. These results highlight the value of scalable real-world data, efficient action generation, and corrective learning for general-purpose VLA policies in deformable-object manipulation.
arXiv:2606.18209v1 Announce Type: new Abstract: Dataset distillation (DD) has emerged as a prominent approach in data centric machine learning, aiming to synthesize compact training sets for efficient training by compressing the information in large datasets into a small number of synthetic samples. However, DD methods are often evaluated under inconsistent evaluation protocols, ranging from standard ERM to single/multi-teacher supervision, making it difficult to isolate the effectiveness of distilled data from evaluation. Moreover, many prior methods claim that DD outperforms data pruning approaches such as coreset selection (CS), based on the assumption that restricting condensed datasets to subsets of real samples fundamentally limits their expressiveness. In this work, we critically evaluate DD methods through large-scale experiments using standardized datasets and evaluation protocols to assess their intrinsic effectiveness. We benchmark seven state-of-the-art (SOTA) DD methods on ImageNet-1K, ImageNet100, and ImageNette, using three widely adopted training protocols against three CS strategies. Our results show that while some DD methods fail to outperform even simple random subsets, the SOTA DD approaches are comparable to or worse than coresets on large-scale datasets and incur a substantially higher cost for construction. Beyond accuracy, we also evaluate the representativeness, diversity, and quality of condensed sets, and find that coresets consistently achieve better coverage of the original data distribution. These findings highlight the limited practical advantages of current DD methods and show that coresets remain competitive and are often a more computationally efficient alternative for data-centric learning.
arXiv:2503.07459v3 Announce Type: replace Abstract: Complex medical reasoning requires integrating heterogeneous clinical evidence across multiple inference steps. Large language models (LLMs) now approach this through two routes: internalized reasoning and externalized agent scaffolding (frameworks that decompose problems collaboratively amongst multiple LLMs). To determine whether these routes are exclusive or complementary, we introduce MedicalAgentsBench, a filtered benchmark of 862 complex clinical questions drawn from the union of eight medical datasets via difficulty-aware curation and contamination screening. Evaluating three internalized reasoning models (DeepSeek-R1, o1-mini, and o3-mini), seven base models, and nine externalized agent-based methods, we find that internalized and externalized approaches each independently improve performance, and that their benefits compound: the highest accuracy is achieved by layering agent workflows onto an internalized reasoning model (i.e., o3-mini + MDAgents with 35.1%). Pareto analysis shows this combination dominates the cost-performance frontier; moreover, lightweight optimization on inexpensive models offers an entry point for resource-constrained settings. Our benchmark is at https://github.com/gersteinlab/MedicalAgentsBench.
arXiv:2606.17229v1 Announce Type: new Abstract: A model that lies while knowing the truth is the central case ELK cannot handle with behavioral evaluation alone. We ask whether such deception leaves an internal signature distinguishing it from honest error. Our key move is a control for wrongness: we contrast a sleeper agent (knows the truth, lies on trigger) against a naive liar (fine-tuned to emit the same wrong answers with no honest training). Both produce identical wrong outputs; any difference is about knowledge conflict, not incorrectness. We find deceptive forward passes carry a conflict signature - 2.1-2.3x higher residual rank than naive-liar passes on the same wrong answer - strong enough to identify which of two responses is the lie with 100% accuracy and no labels, across GPT-2 small/medium (three seeds) and three instruct models. Across Qwen2.5-1.5B/7B and Phi-3-mini, instructed deception raises residual rank on every tested fact (18/18, 40/40, 34/34); on Phi-3, lies separate perfectly from both honest answers and hallucinations (AUC 1.0, Wilcoxon p~6e-11). The signature survives strategic self-constructed deception (model invents its own lie, AUC 1.0), active concealment attempts (AUC 1.0), and length-controlled replication (20/20, AUC 1.0, p~1e-6). Using basis-free relative representations, a probe trained on one model family detects deception in two other families zero-shot (mean AUC 0.933), surviving simultaneous architecture and format change (AUC 0.821), and transfers across five languages (AUC 1.000, length-controlled). The signature is read-only: detectable but not injectable (0/8 both directions). Honest limitations and six negative experiments are documented in full.
arXiv:2606.17635v1 Announce Type: new Abstract: A quantum-enhanced approach for fast temperature diagnostics in ultrashort laser surface processing is introduced. The goal is to overcome limitations of existing methods, such as plasma emission, emissivity changes during ablation, and the need for time-consuming pump-probe measurements. The proposed method exploits polarization anisotropy in entangled photon pairs. The idler photon interacts with the laser-affected material surface, while its entangled counterpart is detected in a remote optical arm. Temperature-dependent changes in the complex refractive index modify the reflectance of p- and s-polarizations on the idler path, altering the coincidence-resolved polarization statistics of the signal photons. Using a Qiskit-based model incorporating experimental pump-probe reflectometry data, remote reconstruction of rapid thermal dynamics during femtosecond laser ablation is demonstrated. Although based on simulation, the model employs literature data to represent realistic material behavior. Due to the discrete nature of single-photon events, classical sliding-window analysis suffers from shot noise and temporal inertia. To overcome this limitation, a multilayer perceptron (MLP) regression network is applied to extract implicit anisotropy information from the photon bitstream. Compared with the classical approach, the neural-network method improves reconstruction robustness, reduces temperature noise, enhances the signal-to-noise ratio (SNR), and enables nanosecond-scale tracking of thermal dynamics. The results indicate that entangled-photon polarization anisotropy combined with machine-learning analysis is a promising approach for remote, background-rejected, high-speed temperature diagnostics in laser-matter interaction studies.
arXiv:2606.17295v1 Announce Type: cross Abstract: The full potential of artificial intelligence in tibial plateau fracture characterisation remains unrealised, constrained by a fundamental dependency on labelled datasets whose consistency cannot be guaranteed: conventional classification schemes such as Schatzker and AO/OTA suffer from inter-observer variability, causing supervised models to learn human disagreement rather than stable fracture morphology. We design, implement, and validate a label-agnostic framework that eliminates this constraint by learning fracture representations directly from imaging data without observer-assigned labels. A RadImageNet-pretrained ResNet-50 encoder is fine-tuned on 154 cleaned knee radiographs using the SimCLR contrastive objective, preceded by a data cleaning protocol and followed by UMAP dimensionality reduction and k-means clustering to discover four imaging-derived phenotypes. Phenotype validity is assessed through a blinded expert review protocol administered to two independent clinicians. The four phenotypes demonstrate robust stability (bootstrap ARI = 0.319 +/- 0.041), strong internal cohesion (silhouette = 0.511), and coherence ratings of 3-5/5 from both reviewers under blinded conditions; one phenotype was unanimously identified as exhibiting comminution -- a high-complexity feature isolated without any supervisory signal. Inter-partition comparison against Schatzker labels yields ARI = 0.013, confirming orthogonality to conventional classification boundaries. Notably, expert reviewers anchored to established classification vocabularies perceived imaging-derived groups as heterogeneous precisely where Schatzker alignment was lowest, suggesting that Schatzker-trained perception and label-agnostic embedding geometry measure orthogonal dimensions. These findings establish label-agnostic SSL phenotyping as a reproducible and clinically interpretable complement to conventional classification.
arXiv:2606.17918v1 Announce Type: new Abstract: We develop and analyze a conforming Virtual Element Method (VEM) for the fourth-order Poisson-Nernst-Planck-Navier-Stokes (PNP-NS) system. The proposed scheme is based on compatible discretizations of each component: an \(H^2\)-conforming VEM for a fourth-order electrostatic potential equation, an \(H^1\)-conforming VEM for the Nernst--Planck equations, and a \textit{divergence-free} and \textit{pressure-robust} VEM for the Navier--Stokes equations. Time integration is performed using a backward Euler scheme to ensure stability. We establish the well-posedness of the continuous problem up to three-dimensions and also establish existence and uniqueness of the fully discrete solution via a fixed-point argument. Further, we derive a priori error estimates showing that the electrostatic potential, concentration and velocity converge optimally in Bochner norms \(L^\infty\bigl(0,T; H^2(\Omega)\bigr)\), \(L^2\bigl(0,T; H^1(\Omega)\bigr)\), and \(L^2\bigl(0,T; \bm{H}^1(\Omega)\bigr)\), respectively. The analysis requires a sophisticated argument to avoid any restrictive assumptions on the coercivity and continuity constants and to handle the trilinear form involving three different variables. The pressure-robust design permits the use of lowest-order pressure approximations without compromising convergence rates of the other variables, reducing computational cost. Numerical experiments confirm the theoretical convergence rates for various polynomial orders and demonstrate the scheme's robustness, including in low-viscosity regimes.
arXiv:2606.18060v1 Announce Type: new Abstract: As Large Language Model based agents enter autonomous scientific research, their ability to resist pseudoscience becomes increasingly important. Otherwise, such systems may rapidly generate plausible yet misleading studies that contaminate academic literature and erode trust in science. We present PseudoBench, an adversarial benchmark for evaluating whether agentic auto-research systems can identify and resist pseudoscientific narratives. PseudoBench contains 200 curated pseudoscientific claim-evidence pairs across five domains and evaluates agents through an end-to-end research pipeline from experiments to writing. Testing seven state-of-the-art agents, we find that current systems readily produce persuasive reports that align with pseudoscientific premises with near-zero refusal rates and the highest resistance of only 27.4%. Stronger agents risk packaging pseudoscience in more sophisticated scientific language, increasing its apparent credibility. These findings reveal an alarming capacity to fuel pseudoscience, calling for scientific alignment before widespread deployment.
arXiv:2606.18203v1 Announce Type: new Abstract: The LLM-empowered personal health agents with user health (sensor) metrics have offered a promising pathway to alleviate global disparities in healthcare access. However, large-scale clinical deployment remains constrained by an open-ended evaluation bottleneck: physician annotation is reliable but costly and unscalable, while LLM-as-a-judge evaluators are scalable but subjective, inconsistent, and sometimes clinically misaligned. We introduce RubricsTree, a scalable evaluation framework with an expert-aligned hierarchical taxonomy of over 100 atomic, clinically-verifiable Boolean rubrics, evolving from the insights of 4,000 real user queries through an iterative human-in-the-loop curation protocol with an expertise panel led by an experienced physician. A context-aware adaptive router activates only the relevant auto-weighted rubric subset per query, providing the throughput needed for scalable evaluation with expert-aligned quality. Through a systematic meta-evaluation, we show that RubricsTree (i) substantially exceeds a strong large-scale evaluation baseline in expert alignment on challenging open-ended queries; (ii) reliably penalizes contextually degraded responses; and (iii) when used as structured instructions, text feedback, or training rewards for performance optimization, yields up to ~66% relative gains on HealthBench for Gemini, GPT, and Qwen model families. RubricsTree thus provides a scalable, auditable, and evolving evaluation infrastructure required for the continuous optimization of product-level personal healthcare AI.
arXiv:2606.17265v1 Announce Type: cross Abstract: The mirror symmetries of a periodic unit cell are exploited to decompose the standing-wave eigenproblem at the high-symmetry vertices of the Brillouin zone into four independent sub-problems on a quarter-cell, each governed by Neumann (sound-hard) or Dirichlet (sound-soft) boundary conditions. Sorting and pairing the resulting eigenfrequencies by index along each segment of the irreducible Brillouin zone boundary yields an explicit formula for the stop-band intervals without computing the full dispersion diagram. The decomposition is exact, following directly from the representation theory of the little group at each high-symmetry point. It applies to any unit cell whose material distribution is invariant under the mirrors normal to the cell faces. The method is validated on two configurations: a phononic crystal of lead cylinders in an epoxy matrix, analyzed using the plane-wave expansion, and a lattice of coupled C-shaped Helmholtz resonators, analyzed using finite-element analysis. For both systems, the reconstructed stop-band boundaries agree with the full Floquet dispersion calculation to within 1% for the lowest bands, requiring eigenvalue solutions at only three discrete wavevectors. Avoided crossings within a Brillouin zone segment can cause bands to exhibit non-monotone behavior, rendering the pairing rule approximate; the spectral conditions for this are identified. Flat bands common to both boundary-condition types are identified as bound states in the continuum.
Bayesian Poisson-Randomized Gamma Tensor Factorization with Application to International Trade Flows
arXiv:2606.17267v1 Announce Type: cross Abstract: We study sparse semi-continuous tensor data with excess zeros, heavy right tails, and slice-specific dispersion. Such features arise naturally in monetary-valued multi-way data, such as international trade, where most exporter--importer--product--year cells are zero while positive values are continuous and highly variable. To model these data, we propose a Bayesian hierarchical tensor factorization model that places a low-rank CP structure on a latent Poisson rate tensor and couples it with a conditional Gamma model for positive outcomes, with rate parameters that can vary across slices within a mode. The model therefore separates the occurrence and magnitude of positive observations while borrowing strength across all tensor dimensions through a shared low-rank latent structure. To scale posterior inference to large arrays, we develop a hybrid variational--Monte Carlo algorithm that combines efficient coordinate ascent updates with a partially collapsed augmented-data sampler. Applied to approximately 60 million trade flows, the method surfaces multiway dependence across exporters, importers, products, and years that is difficult to recover from gravity-type or pairwise network analyses, which do not jointly model the product and temporal dimensions.
arXiv:2606.17268v1 Announce Type: cross Abstract: Generalizing the construction of two-block group algebra (2BGA) codes, we introduce a family of two-block quantum LDPC codes constructed using the action of a group on the cosets of its subgroup. This replaces the regular group actions of the earlier two-block constructions and significantly expands the search space, yielding new quantum LDPC codes outside the 2BGA family. Through a computer search, we identify several new quantum LDPC codes, including weight-6 codes with parameters $[[48,8,6]]$, $[[96,8,10]]$, and $[[224,12,16]]$, as well as weight-8 codes with parameters $[[84,16,8]]$, $[[112,16,10]]$, $[[128,16,12]]$, and $[[168,16,15]]$. Furthermore, we introduce a maximally packed syndrome extraction schedule of depth $w+2$, including initialization and measurement steps, for any code with a maximum stabilizer weight of $w$ from our family. Under a standard circuit-level noise model, our codes, when decoded using BP-OSD, perform competitively with BB codes, achieving thresholds of $\approx0.65\%$ for the weight-6 family and $\approx0.35\%$ for the weight-8 family. Finally, we introduce a group-theoretic framework to generate sequences of graph-based covers of 2BGA codes, recovering and extending recent results on code constructions of this type.
arXiv:2606.18200v1 Announce Type: new Abstract: Learned PDE simulators are increasingly used as low-cost replacements for expensive numerical solvers, but standard relative $L^2$ error does not determine whether a learned model behaves as a coherent numerical time propagator. This paper presents a diagnostic software suite for auditing learned PDE simulators as approximate evolution operators. The suite provides architecture-independent, post hoc diagnostics for relative state error, semigroup consistency, finite-difference generator discrepancy, energy behavior, integral balance, admissibility constraints, perturbation response, and scaling-law consistency. The software is designed around a minimal contract: reference trajectories, a learned propagator or saved predictions, equation metadata, and a diagnostic configuration specifying which structures are meaningful for the problem under study. We validate the suite on five benchmark PDE tasks: two-dimensional incompressible Navier-Stokes, shallow-water dynamics, active matter, three-dimensional compressible Navier-Stokes, and three-dimensional magnetohydrodynamics, using FNO, DeepONet, U-Net, and ResNet-style surrogate models together with controlled underfit and oversmoothed variants. The validation study shows that relative $L^2$ error can remain moderate, or even improve, while structural diagnostics deteriorate substantially. The package therefore supports software-level auditing of learned PDE simulators by reporting an interpretable diagnostic panel rather than collapsing model behavior into a single state-error score.
arXiv:2606.18215v1 Announce Type: new Abstract: We present PhysiBench, an open resource for developing and evaluating computational methods in systems biology including a benchmark suite of 612 executable intracellular Boolean regulatory network variants and a dataset of 120,000 time-resolved multiscale stochastic simulations. The benchmark models are derived from seven published Boolean networks spanning cell-cycle control, developmental patterning, cancer signaling, immune response, and cell-fate decisions, and are executable in the PhysiBoSS/PhysiCell multiscale simulation framework. Model variants are generated through mutation-based model construction, online behavioral filtering, and offline sensitivity evaluation. The simulation dataset is produced from 60 selected models under systematically sampled stimulation protocols and fixed model-level initial configurations. Each trajectory is linked to its model identifier, input-parameter file, stochastic seed, and cell-level output file. PhysiBench supports direct simulation, surrogate modeling, data-driven inference, simulation-based optimization, and comparative benchmarking. Technical validation includes file-integrity and executability checks, graph-based structural diversity analyses, and behavioral heterogeneity assessment from multiscale simulation outputs.
arXiv:2606.18220v1 Announce Type: new Abstract: Consensus protocols form the core of blockchains and other replicated state machines, ensuring that all correct nodes process the same totally ordered log of input transactions. In fault-free executions, performance is driven by the good-case transaction latency -- the time between a transaction becoming known to all nodes and its confirmation by the consensus protocol -- which depends on both how frequently proposals are made and, once made, how quickly they are confirmed. While prior work has established tight lower bounds on confirmation latency that modern protocols already achieve, it remains open whether the inter-proposal time can be further reduced below the state-of-the-art of one network delay. We introduce Gatling, an atomic broadcast protocol that achieves arbitrarily small inter-proposal times under rotating leader schedules; in particular, smaller than the network delay. Gatling runs multiple parallel instances of a black-box atomic broadcast protocol and staggers their proposal schedules to generate proposals in faster succession than state-of-the-art protocols. A deterministic interleaving rule merges the outputs of these instances into a single global log. We analyze the effects of head-of-line blocking caused by crashed leaders, and derive Gatling's optimal number of parallel instances. We further study the impact of Gatling on predictable validity and present two variants that retain this property. Finally, our experiments confirm that Gatling can be used with off-the-shelf component protocols to achieve low latency without fine-tuning the component protocol for minimum latency.
arXiv:2509.21886v3 Announce Type: replace Abstract: Learning to compute, the ability to model the functional behavior of a circuit graph, is a fundamental challenge for graph representation learning. Yet, the dominant paradigm is architecturally mismatched for this task. This flawed assumption, central to mainstream message passing neural networks (MPNNs) and their conventional Transformer-based counterparts, prevents models from capturing the position-aware, hierarchical nature of computation. To resolve this, we introduce TRACE, a new paradigm built on an architecturally sound backbone and a principled learning objective. First, TRACE employs a Hierarchical Transformer that mirrors the step-by-step flow of computation, providing a faithful architectural backbone that replaces the flawed permutation-invariant aggregation. Second, we introduce function shift learning, a novel objective that decouples the learning problem. Instead of predicting the complex global function directly, our model is trained to predict only the function shift, the discrepancy between the true global function and a simple local approximation that assumes input independence. We validate this paradigm on various circuits modalities, including Register Transfer Level graphs, And-Inverter Graphs and post-mapping netlists. Across a comprehensive suite of benchmarks, TRACE substantially outperforms all prior architectures. These results demonstrate that our architecturally-aligned backbone and decoupled learning objective form a more robust paradigm for the fundamental challenge of learning the functional behavior of a circuit graph.
arXiv:2602.23116v3 Announce Type: replace Abstract: We consider the problem of regularized best-response max-regret minimization in online RLHF under general preferences and bandit feedback. While various regularizers are utilized to robustify alignment, known polylogarithmic regret guarantees remain heavily specific to KL. To investigate whether such fast rates extend beyond KL, we adopt the Generalized Bilinear Preference Model (GBPM) -- capturing intransitive preferences over $d$-dimensional item-wise features via a rank-$2r$ skew-symmetric matrix -- to isolate the impact of generic regularization. Crucially, under GBPM, we prove that the dual gap of any greedy policy is bounded by the squared estimation error, derived using \emph{only} strong convexity and skew-symmetry. Under a feature coverage assumption, we establish a \emph{generic} polylogarithmic regret of $\tilde{\mathcal{O}}(\eta d^4 C_{\min}^{-1} (\log T)^2 \wedge d^2 C_{\min}^{-1/2} \sqrt{T})$ with Greedy Sampling, and a dimension-wise improved regret (for well-conditioned arm-sets) of $\tilde{\mathcal{O}}(C_{\min}^{-2} \sqrt{\eta r T} \wedge r^{1/3} C_{\min}^{-4/3} T^{2/3})$ with Explore-Then-Commit, where $\eta^{-1}$ is the regularization coefficient, $T$ is the time horizon, and $C_{\min}$ is an arm-set dependent quantity. This demonstrates that ``fast'' regrets are not KL-specific, but rather a fundamental consequence of generic strongly convex geometry.
arXiv:2606.17389v1 Announce Type: new Abstract: Multimodal Foundation Models are increasingly used as reasoning agents, making reliability, knowing when a model may hallucinate, critical. A common intuition, which we call the Attention-Confidence Assumption, holds that reliability follows from "structural" visual perception: tight attention on relevant regions should signal a trustworthy answer, while scattered attention signals confusion. We challenge this through the VLM Reliability Probe (VRP), a systematic cross-family study of reliability signals in contemporary Vision-Language Models (VLMs). We introduce structural-attention metrics, cluster counts (C_k) and spatial entropy (H_s), to quantify the visual encoder's gaze, and track its evolution (Delta H_s) across layers. This reveals a "Symbolic Detachment": models often "Early Lock" visual features only to diffuse attention later, severing early perception from final generation. Contrary to the grounding hypothesis, we find a "Cluster Failure": spatial attention has near-zero correlation (R approx 0.001) with accuracy. Instead, reliability is a phenomenon of generation dynamics and internal-state distributions. Self-Consistency, the agreement rate across sampled reasoning paths, is the dominant predictor of truth (R = 0.429). Scaling causal interventions exposes a sharp architectural divergence: LLaVA locks its prediction in a fragile late-stage bottleneck, whereas PaliGemma and Qwen2-VL distribute reliability globally, staying resilient even when ~50% or more of their most predictive layer is destroyed. For current VLMs, reliability signals are detached from visual grounding maps and are best inferred from generation-time dynamics and hidden-state probes.