arXiv:2601.12282v2 Announce Type: replace Abstract: The functions of different regions of the human brain are closely linked to their distinct cytoarchitecture, which is defined by the spatial arrangement and morphology of the cells. Identifying brain regions by their cytoarchitecture enables various scientific analyses of the brain. However, delineating these areas manually in brain histological sections is time-consuming and requires specialized knowledge. An automated approach is necessary to minimize the effort needed from human experts. To address this, we propose CytoCLIP, a suite of vision-language models derived from pre-trained Contrastive Language-Image Pre-Training (CLIP) frameworks to learn joint visual-text representations of brain cytoarchitecture. CytoCLIP comprises two model variants: one is trained using low-resolution whole-region images to understand the overall cytoarchitectural pattern of an area, and the other is trained on high-resolution image tiles for detailed cellular-level representation. The training dataset is created from NISSL-stained histological sections of developing fetal brains of different gestational weeks. It includes 86 distinct regions for low-resolution images and 379 brain regions for high-resolution tiles. We evaluate the model's understanding of the cytoarchitecture and generalization ability using region classification and cross-modal retrieval tasks. Multiple experiments are performed under various data setups, including data from samples of different ages and sectioning planes. Experimental results demonstrate that CytoCLIP outperforms existing methods. It achieves a weighted F1 score of 0.87 for whole-region classification and 0.91 for high-resolution image tile classification.
Science Journals
arXiv:2606.29757v1 Announce Type: new Abstract: Standard stereo VIO focuses exclusively on the benefit of metric scale via single-spectrum baselines, often overlooking the risks of spectral redundancy. This structural limitation leads to correlated failures, where both sensors simultaneously fail in degraded environments that affect their shared spectrum. Leveraging a cross-spectral system presents a complementary solution to this issue, yet the significant appearance gap between modalities renders standard matching ineffective. Existing deep learning-based matchers, while effective, introduce inference latencies that violate real-time constraints. To bridge this gap, we present an asynchronous real-time cross-spectral visual-thermal-inertial (VTI) system that temporally decouples high-latency deep matching from high-rate state estimation. Our architecture incorporates a spectral-aware weighting scheme that dynamically balances modality reliance based on photometric entropy and thermal noise, ensuring robustness against both abrupt lighting changes and thermal artifacts. Furthermore, we introduce a seamless handling mechanism for thermal Non-uniformity Correction (NUC) to maintain tracking continuity. Extensive experiments across diverse scenarios confirm that our system overcomes spectral redundancy, yielding superior accuracy in nominal daylight while ensuring robustness in visually degraded environments. We will open source our code and data: https://github.com/seungsang07/cross-spectral-stereo-inertial-odometry
arXiv:2606.29763v1 Announce Type: new Abstract: Topological data analysis (TDA), particularly persistent homology (PH), captures geometric structural properties in medical images (e.g., connected components, loops, shape characteristics), which conventional pixel-level deep learning approaches often neglect. While many topological descriptors are known for converting persistence diagrams (PDs) or raw images into topological feature vectors, existing methods mostly default to a single fixed descriptor (e.g., persistence images), leaving the diversity of topological representations largely unexplored. To the best of our knowledge, there is no known large language model (LLM)-based agentic framework that can automatically determine the most suitable topological descriptors for a given image dataset and produce the corresponding topological feature vectors for downstream tasks. To fill this gap, we propose \textbf{TopoAgent}, an LLM-based agentic framework that automates topology learning for medical image analysis.TopoAgent operates through a Perception--Reasoning--Action--Reflection loop supported by 21 domain-specific tools and dual memory that accumulates experience across runs. Its skill set is distilled from systematic evaluation of 15 topological descriptors across 26 datasets with six classifiers. TopoAgent analyzes input images and their topological characteristics, reasons about which topological descriptors best suit the input, and determines the optimal descriptor and its configuration, all without task-specific training.
arXiv:2606.28942v1 Announce Type: new Abstract: A deep-etch field termination strategy using a Ga-assisted plasma-free etching technique in a low-pressure chemical vapor deposition (LPCVD) system is demonstrated for $\beta$-Ga$_2$O$_3$ Schottky barrier diodes (SBDs). The thermally activated etching method provides a plasma-free approach for forming deep mesa terminations while maintaining excellent device integrity. The fabricated diodes exhibit excellent forward conduction characteristics with a turn-on voltage of 1.14~V, a Schottky barrier height (SBH) of 1.15~eV, an ideality factor of 1.20, and a specific on-resistance of 3.72~m$\Omega\cdot$cm$^2$, all closely matching those of the unetched planar devices. Capacitance-voltage analysis further confirms a uniform carrier concentration of $2\times10^{16}$~cm$^{-3}$ and an SBH of 1.23~eV, indicating stable electrical characteristics after deep mesa formation. Temperature-dependent electrical measurements from 25 to 250$^\circ$C demonstrate stable thermionic-emission transport behavior, with a gradual increase in on-resistance at elevated temperatures due to phonon-limited carrier mobility. Over this temperature range, the SBH decreases from 1.16 to 1.12~eV, while the ideality factor increases from 1.21 to 1.33. The leakage current remains low throughout the entire temperature range, and the rectification ratio remains above $10^{5}$ even at 250$^\circ$C. Under reverse bias, the diodes exhibit an increase in breakdown voltage from 287V to 500V, confirming the effectiveness of geometric electric-field redistribution achieved by the deep-etched mesa structure. Silvaco TCAD simulations corroborate these experimental observations by showing significant suppression of electric-field crowding near the anode edge. These results establish Ga-assisted plasma-free etching as a reliable, damage-free field termination technique for high-performance $\beta$-Ga$_2$O$_3$ power devices.
arXiv:2606.09620v2 Announce Type: replace Abstract: Planning collision-free motion for large robot fleets is difficult because collision avoidance induces strong inter-agent coupling that grows rapidly with team size. We consider omnidirectional floating robots on water, where choreographies are specified by sparse keyframes and an interactive tool must generate trajectories within seconds, even when transitions span minutes and thousands of time steps. We propose a scalable pipeline that builds a collision graph from an initialization, decomposes the coupled problem into interaction clusters, and solves clusters independently (and in parallel) with robustness mechanisms for common decomposition pathologies. We validate the approach in simulations up to 500 robots. The synthesized trajectories have also been deployed in two real-world demonstrations, on Lake Z\"urich with a fleet of 24 Way of Water crafts and at the Time Space Existence 2025 Venice Biennale.
arXiv:2602.21343v2 Announce Type: replace Abstract: Decentralized Federated Learning (DFL) removes the central aggregator of conventional Federated Learning, but peer-to-peer model exchange still exposes network traces: who communicates, when fragments move, and which packets correlate across rounds. This paper studies network-layer sender--message linkability for DFL model sharing and presents UnlinkableDFL, a framework in which every participant acts as both a learner and a peer-based mix relay. Shareable model states are split into uniform, onion-encrypted fragment packets and carried over a peer-run mixnet with cover traffic, randomized delays, and independently sampled multi-hop paths. Nodes then perform fragmented aggregation over local and received fragments without sender identities. The analysis bounds sender-linking probability through route uncertainty and relay shuffles, and characterizes when fragment-level aggregation preserves FedAvg-style behavior. A prototype implements QUIC transport, Sphinx-style packets, and Single-Use Reply Block (SURB) acknowledgments. Experiments show that the design sustains learning under sparse deployment while exposing a privacy--cost trade-off: path diversity and relay mixing raise network-layer uncertainty, whereas delay and forwarding dominate overhead. Stress tests confirm robustness to churn and Byzantine updates. A curious-recipient attack marks the boundary of the network-layer guarantee, where payload-level fingerprints survive network-layer anonymization and need complementary defenses, although partial updates and more IID data weaken this attack surface.
arXiv:2602.21567v2 Announce Type: replace Abstract: While the rapid proliferation of electric vehicles (EVs) accelerates net-zero goals, uncoordinated charging activities impose severe operational challenges on distribution grids, including exacerbated peak loads, thermal overloading, and voltage violations. To overcome the computational intractability of jointly optimizing grid infrastructure reinforcements and Battery Energy Storage System (BESS) installations, this paper proposes a novel three-stage Diagnosis-Driven Co-Planning (DDCP) framework. The methodology integrates a Violation Detection and Quantification (VDQ) model to systematically identify system breaches, and a Violation Mitigation-Based Planning (VMBP) model for optimal BESS allocation. Specifically, Stage I of the DDCP framework diagnoses critical bottleneck lines that render standalone BESS solutions infeasible; Stage II executes targeted physical upgrades exclusively on these bottlenecks; and Stage III finalizes the optimal BESS deployment on the updated network topology. Furthermore, this study quantifies the EV hosting capacity thresholds before and after BESS integration across varying EV adoption rates and base voltages. Finally, a comprehensive comparative analysis evaluates four mitigation approaches: the VDQ-driven cable upgrade (VCU) model, the VMBP model, system-wide voltage uprating, and the proposed DDCP framework. The results demonstrate that the DDCP framework not only resolves the complex joint-optimization hurdle but also achieves superior techno-economic performance in addressing high-EV-penetration challenges.
arXiv:2606.29941v1 Announce Type: new Abstract: Visuo-Tactile policies leveraging optical tactile sensors have shown great promise in contact-rich manipulation. These sensors achieve high spatial resolution and multi-dimensional force sensing by utilizing an internal camera to monitor the deformation of their elastic gel surface, thereby indirectly inferring tactile cues. Despite their advantages, extracting fine-grained contact states necessary for contact-rich manipulation remains an open challenge. Existing methods typically use either raw images or cumulative motion fields to represent tactile cues. However, both are prone to perception ambiguity. Raw tactile images mainly capture appearance changes, while cumulative motion fields only reflect the aggregate gel deformation. Consequently, distinct fine-grained contact states can exhibit highly similar patterns, making it difficult to explicitly distinguish subtle contact variations. To address this issue, we explore the dynamic priors of tactile motion and discover that the correlation between transient and cumulative motion can explicitly distinguish fine-grained contact states. Based on this insight, we propose a motion-aware tactile representation to facilitate contact-rich manipulation. Beyond tactile representation, effective fusion of tactile and visual modalities is also critical. Most existing fusion methods either directly concatenate features from each modality or train modality-specific networks separately and fuse their outputs. However, these strategies struggle to simultaneously model cross-modal interactions and preserve modality-specific characteristics. In this work, we take advantage of the Mixture-of-Transformers architecture and propose a unified modality-aware visuo-tactile policy that captures cross-modal complementarity while maintaining modality-specific properties.
arXiv:2605.00994v2 Announce Type: replace Abstract: Finetuning can significantly modify the behavior of large language models, including introducing harmful or unsafe behaviors. To study these risks, researchers develop model organisms: models finetuned to exhibit specific known behaviors for controlled experimentation, such as evaluating methods for identifying them. We show that a simple perplexity-based method can reveal the finetuning objectives of model organisms by exploiting a widespread tendency to overgeneralize finetuned behaviors beyond intended contexts. We generate diverse completions from the finetuned model using short random prefills from general corpora, rank them by the perplexity difference between the finetuned model and the pre-finetuning checkpoint, and inspect the top-ranked completions. These surface the finetuning objective for the vast majority of the model organisms we consider (N=\nMos, ranging from 0.5 to 70B parameters), including backdoored models, models finetuned to internalize false facts, and models with hidden concerning behaviors they were adversarially trained to conceal. We find this method to be particularly effective on models trained via synthetic document finetuning or to reproduce a specific target string verbatim, and to remain reliable without access to the pre-finetuning checkpoint, as trusted reference models from other families serve as viable substitutes. Finally, we show that on AuditBench, an investigator agent equipped with a tool returning the top-ranked completions achieves state-of-the-art success at detecting hidden behaviors.
arXiv:2606.16364v2 Announce Type: replace Abstract: LLM agents mis-call tools, and the natural guess is that the model failed to see the right tool in a crowded harness. We show the opposite through a lens concurrent work sets aside -- the model's attention to labeled tool-definition segments. On real BFCL failures, by per-candidate attention argmax the model attends most to the correct tool 80% of the time (vs. 21% chance), and the gold is the under-attended segment on only 10%: it looks at the right tool and still picks wrong. This directly refutes the intuitive "crowded-harness / lost-in-the-middle" explanation: the failure is at the decision readout, not the harness, and we pin it there three ways. (1) Input vs. readout: repairing the prompt (reordering or duplicating the gold tool) recovers <=23% of failures, while readout-side interventions recover 59-91%. (2) Representation-invariance: two gold-pointed interventions in different representations -- an additive attention-logit bias and a residual-stream steering vector -- recover largely the same failures (per-task Jaccard 0.865 pooled, 0.79-0.91 per model), so the bottleneck is localized to the readout independent of which representation is poked. (3) A training-free, gold-free selector: per-segment attention closes most of the gold-free-vs-oracle gap on BFCL (+11.9 pts pooled function-name selection vs. +17.9-pt oracle headroom) and adds +14.9 pts on Seal-Tools; every model positive (exact McNemar p<=8e-4 each). Scopes differ: the causal attention-bias dose-response is bidirectional and monotonic on 10 mask-honoring models (3-32B), the full 0.5-32B span carrying only the correlational diagnostic; the deployable selector is evaluated on 5 single-turn models and does not yet transfer to a multi-turn loop.
arXiv:2606.30067v1 Announce Type: new Abstract: We introduce Neural Subspace Reallocation (NSR), which reframes continual learning as memory management over parameter subspaces. Instead of treating Low-Rank Adaptation (LoRA) modules as disposable per-task adapters, NSR manages them as compressible, retrievable memory units on a frozen backbone through a recurring cycle: (1) compress learned LoRAs via SVD, (2) reserve them in a TaskKnowledgeBank, (3) recall related past LoRAs by embedding similarity to warm-start new or returning tasks, and (4) reallocate the active subspace accordingly, with distillation protecting prior tasks. We prove that in cyclic environments any memoryless allocation policy incurs cumulative regret Omega(T(M-1)Delta_switch) relative to a history-aware policy backed by the Bank (Theorem 1). Empirically, on Split-CIFAR-100 the Bank reduces cyclic recovery time by 10x, exactly as predicted, and on the heterogeneous 5-Datasets benchmark NSR achieves the highest accuracy and the least forgetting, about 9x closer to zero backward transfer than the memoryless heuristics. Crucially, we run a controlled study that isolates which component matters: holding the Bank fixed and varying only the allocation rule, we find that a simple similarity-based retrieval rule matches or beats a learned reinforcement-learning controller (recovering recurring tasks in 0 vs 1.8 steps and reaching equal accuracy). Our central, honest finding is therefore that the memory mechanism -- compression and similarity retrieval -- rather than a learned allocation policy, drives continual-learning performance under fixed capacity. A memory-budget analysis confirms the compressed Bank stays small -- 0.29 MB of parameter memory per task -- so a top-K retention cap bounds the total footprint while preserving fast recovery for retained tasks.
arXiv:2606.29222v1 Announce Type: new Abstract: Autonomous navigation in unknown environments requires a robot to efficiently reach a predefined goal while exploring without prior maps. Although progress has been made in this area, most existing works still rely on traditional planning methods with hand-crafted rules, while learning-based methods often suffer from limited environmental memory and challenges in simulation-to-real (sim-to-real) transfer. To overcome these limitations, we propose a Contextual-memory Oriented Reinforcement-learning (CORE) planner for robot navigation in unknown environments. The proposed CORE planner effectively combines the core advantages of traditional and learning-based methods. Specifically, our method uses a sparse visibility graph for structured environment representation, reducing the computational overhead of dense grid maps, and employs a Transformer network to achieve a holistic environmental understanding, thereby significantly improving navigation efficiency. Moreover, we introduce a visibility graph-based graph sparsification method and a contextual memory mechanism, which alleviates local optima and enhances computational performance in large-scale scenes. Finally, our approach achieves zero-shot sim-to-real transfer after training solely on image-based environments, requiring no fine-tuning. Experimental results show that CORE Planner consistently outperforms state-of-the-art methods, including the traditional FAR Planner and all learning-based baselines, across representative environments, reducing travel distance by 13\% over traditional FAR Planner and by up to 48\% relative to learning-based baselines, with larger gains observed in more complex environments. In real-world scenarios, CORE successfully navigates without human intervention, showcasing zero-shot sim-to-real transfer. Code is available at https://github.com/BBD00/core_planner.
arXiv:2507.00719v3 Announce Type: replace Abstract: Typically, numerical simulations of Earth systems are coarse, and Earth observations are sparse and gappy. We apply four generative diffusion modeling approaches to super-resolution and inference of forced two-dimensional quasi-geostrophic turbulence on the beta-plane from coarse, sparse, and gappy observations. Two guided approaches minimally adapt a pre-trained unconditional model: SDEdit modifies the initial condition, and Diffusion Posterior Sampling (DPS) modifies the reverse diffusion process score. Two conditional approaches, a vanilla variant and classifier-free guidance, require training with paired high-resolution and observation data. We consider multiple test cases spanning: two regimes, eddy and anisotropic-jet turbulence; two Reynolds numbers, 10^3 and 10^4; and two observation types, 4x coarse-resolution fields and coarse, sparse and gappy observations. Our comprehensive skill metrics include norms of the reconstructed vorticity fields, turbulence statistical quantities, and quantifications of the super-resolved probabilistic ensembles and their errors. We also study the sensitivity to tuning parameters such as guidance strength. Results show that the generated super-resolution fields of SDEdit are unphysical, while those of DPS are reasonable but with smoothed fine-scale features; however, neither of these lower-cost models propagates observational information effectively to unobserved regions. The two conditional models require re-training, but reconstruct missing fine-scale features, are cycle-consistent with observations, and predict correct turbulence statistics, including the tails. Further, their mean errors are highly correlated with and predictable from their ensemble standard deviations. Results highlight the tradeoffs between ease of implementation, fidelity (sharpness), and cycle-consistency of the diffusion models, and offer practical guidance for deployment.
arXiv:2606.29247v1 Announce Type: new Abstract: Vision-Language-Action (VLA) models represent a promising direction for embodied intelligence in surgical robotics. Despite the prevalence of VLA benchmarks for general robotics, standardized evaluation platforms specifically designed for surgical contexts remain absent. To address this limitation, we present SurgVLA-Bench, the first comprehensive benchmark for evaluating VLA models in laparoscopic surgical robotics. Leveraging the SurRoL simulation platform, we construct a hierarchical task taxonomy ranging from atomic actions to complete surgical procedures, complemented by a multi-dimensional evaluation framework assessing action accuracy and semantic consistency. We then systematically evaluate two representative paradigms, including autoregressive models such as OpenVLA, and flow matching models such as $\pi_{0}$, $\pi_{0.5}$, and SmolVLA. Our experiments show that autoregressive models tend to excel in semantic understanding, while flow matching models often achieve higher task precision but may face generalization trade-offs. However, even the best-performing models remain far from satisfactory, as the constrained endoscopic field of view, restricted viewing angles, and frequent occlusions persist as fundamental physical bottlenecks. The code and data are available at https://github.com/VCL-HNU/SurgVLA
arXiv:2606.29252v1 Announce Type: new Abstract: We study repeated bidding in multi-unit discriminatory (pay-as-bid) auctions for a single bidder with per-round utility equal to value minus $\alpha$ times payment, where $\alpha\in[0,1]$ is a cost-of-capital parameter. The bidder aims to maximize cumulative utility over $T$ rounds subject to a total budget $B$. The problem is challenging even without budgets: the action space is exponential in $M$, the maximum demand of the bidder and the valuation vector (context) varies over time. Exploiting a decomposition of utility across units, we develop polynomial-time learning algorithms based on shortest paths in a directed acyclic graph, obtaining sublinear regret under both full-information and bandit feedback. In the bandit setting, the regret is independent of the number of contexts due to complete cross-learning: observing the utility of the chosen action under the realized context reveals the utility for the same action under all counterfactual contexts. With budget constraints, when the average normalized per-round budget $\rho=\frac{B}{MT}<1$, we design a coupled primal-dual algorithm in which the DAG-based procedure uses dual-adjusted edge weights for primal updates, while online gradient descent updates the dual variable, yielding $\rho$-approximate sublinear regret. Finally, we give implementations whose per-round time and space are independent of the number of contexts, enabling scalability to large or even infinite context spaces.
arXiv:2508.19094v3 Announce Type: replace Abstract: Event cameras are a bio-inspired class of sensors that asynchronously measure per-pixel intensity changes. Under fixed illumination conditions in static or low-motion scenes, rigidly mounted event cameras are unable to generate any events and become unsuitable for most computer vision tasks. To address this limitation, recent work has investigated motion-induced event stimulation, which often requires complex hardware or additional optical components. In contrast, we introduce a lightweight approach to sustain persistent event generation by employing a simple rotating unbalanced mass to induce periodic vibrational motion. This is combined with a motion-compensation pipeline that removes the injected motion and yields clean, motion-corrected events for downstream perception tasks. We develop a hardware prototype to demonstrate our approach and evaluate it on real-world datasets. Our method reliably recovers motion parameters and improves both image reconstruction and edge detection compared to event-based sensing without motion induction.
arXiv:2606.29414v1 Announce Type: new Abstract: Periodic Implicit Neural Representations (INRs) such as SIREN and FINER assign every neuron, the same global frequency, spending the representational budget inefficiently when local signal content varies. We introduce FiRe (Frequency Reparameterization), that accelerates optimization by reparameterizing per-neuron frequency of periodic INRs without changing their underlying activation function. FiRe gives each neuron a bounded, input-dependent frequency via a separate low-rank gating path and is applicable to any periodic activation function. The gate acts as an implicit preconditioner that improves optimization conditioning at initialization via the Neural Tangent Kernel (NTK). This better-conditioned initialization makes optimization converge faster, and the high-frequency content of the reconstruction tracks the target more closely at a fixed computational budget. On 2D image fitting, FiRe increases PSNR over a parameter-matched baseline (up to +1 dB at short training budgets), with gains that vary with resolution and diminish at full convergence. We characterize how performance depends on resolution, rank, and training budget, and give an NTK account that predicts these trends.
arXiv:2510.19465v2 Announce Type: replace Abstract: Obtaining truly representative pore-scale images that match bulk formation properties remains a fundamental challenge in subsurface characterization, as natural spatial heterogeneity causes extracted sub-images to deviate significantly from core-measured values. This challenge is compounded by data scarcity, where physical samples are only available at sparse well locations. This study presents a multi-conditional Generative Adversarial Network (cGAN) framework that generates representative pore-scale images with precisely controlled properties. The framework was trained on thin section samples from four depths (1879.50-1943.50 m) of a carbonate formation, simultaneously conditioning on porosity and depth within a single model. It processes RGB thin section images that preserve critical mineralogical information (anhydrite-dolomite differentiation, grain boundaries, porosity distinctions) lost in conventional grayscale representations, capturing characteristics from grainstone fabrics to crystalline textures with anhydrite inclusions. The model achieved strong porosity control (R^2 = 0.95) across all formations with mean absolute errors of 0.0099-0.0197. Morphological validation confirmed preservation of average pore radius, specific surface area, and tortuosity within acceptable tolerances. Two-point correlation (S2) analysis confirmed that generated images preserve the spatial continuity and characteristic length scales of natural pore networks, with results consistent across the imaging resolutions tested (1.8-3.0 micron/pixel). Validated against core sample properties, generated images showed higher property fidelity with dual-constraint errors of 1.9-12.4% compared to 37.5-713.6% for randomly extracted real sub-images. This capability provides practical tools for subsurface characterization, particularly valuable for carbon storage, geothermal energy, and groundwater management.
arXiv:2606.30269v1 Announce Type: new Abstract: We study the flow in a symmetric three-dimensional confined sudden expansion with lateral inflow at Reynolds number below 300 and varying lateral-to-central flow rate ratio, using experiments, linear stability analysis, weakly nonlinear theory, and direct numerical simulations. Three distinct flow regimes are identified. Outside an intermediate band of lateral-to-central flow rate ratio, the flow undergoes a steady symmetry-breaking bifurcation above a critical Reynolds number, deflecting the central jet toward one side wall; weakly nonlinear analysis shows this bifurcation to be supercritical, excepting a very narrow parametric range. Within the intermediate band, no such critical Reynolds number exists and direct numerical simulations confirm that residual velocity asymmetries reflect the imposed geometric imperfections rather than intrinsic amplification. Fluctuations observed experimentally in the intermediate band of lateral-to-central flow rate ratio remain unexplained and warrant further investigation.
arXiv:2606.30092v1 Announce Type: new Abstract: Real-time strategy (RTS) games present significant AI challenges, characterized by expansive state-action spaces arising from multi-unit coordination in continuous battlefields, and sparse delayed rewards stemming from final win/lose signals. Existing approaches face a trade-off between managing the dimensionality explosion of joint actions and maintaining the interpretability of complex state representations. This complexity is further intensified by the limitation of traditional hierarchical structures in adaptively decomposing tasks into effective tactical modules. Such difficulties are compounded by the black-box nature of deep learning models and their reliance on sparse rewards, which together result in limited sample efficiency and a lack of decision-making transparency. To address these limitations, this paper proposes HRL-IM/CBS, a hierarchical reinforcement learning framework with influence map hashing and cluster-based scripts for StarCraft micromanagement. Influence map hashing encodes global battlefield situations into compact hexadecimal codes, capturing spatial control and relative advantage. Cluster-based scripts enable dynamic local coordination through adaptive unit partitioning. The hierarchical multi-Q-table architecture decomposes decision-making into upper-level clustering strategy selection and lower-level tactical execution, with reward allocation providing dense learning signals. Experiments across six asymmetric scenarios demonstrate competitive performance against deep RL baselines while offering advantages in sample efficiency and interpretability through transparent Q-table representations.
arXiv:2605.01122v2 Announce Type: replace Abstract: Iterative ptychographic reconstruction algorithms are widely used for coherent diffractive imaging but can exhibit slow convergence under realistic experimental conditions. We propose a machine learning-augmented approach that accelerates iterative ptychographic reconstruction by introducing a learned fast-forward operator applied during reconstruction. Following an initial warm-up using standard iterations, the fast-forward operator advances the reconstruction toward a more converged state, after which conventional iterative updates are resumed. This strategy preserves the physical consistency and flexibility of established ptychographic solvers while reducing the number of iterations required for convergence. The model is trained on diverse ptychographic datasets and evaluated on experimental data acquired in a different year, demonstrating robustness and temporal generalization. Compared with conventional iterative solvers, the machine learning-augmented method achieves comparable reconstruction quality while converging faster in terms of Poisson negative log-likelihood, yielding over a two-fold reduction in wall-clock time. The approach has been integrated into an existing reconstruction pipeline and deployed in production at a synchrotron beamline, demonstrating practicality for real-time experimental operation.
arXiv:2511.21274v3 Announce Type: replace-cross Abstract: This paper proposes a novel analytical framework, denoted the Multiport Analytical Pixel Electromagnetic Simulator (MAPES). MAPES enables efficient and accurate prediction of the electromagnetic (EM) performance of arbitrary pixel-based microwave (MW) and RFIC structures. Unlike the Internal Multiport Method (IMPM), which optimizes only connecting elements within a fixed, gap-separated pixel skeleton, MAPES operates directly on the all-pixel presence/absence formulation used in recent MW/RFIC design. This is enabled by diagonal virtual pixels, an occupancy-to-load mapping, and a multi-layer/via port-level formulation that have no counterpart in IMPM. By introducing virtual pixels and diagonal virtual pixels and inserting virtual ports at critical positions, MAPES captures all horizontal, vertical, and diagonal electromagnetic couplings within a single multiport impedance matrix. Only a small set of full-wave simulations (typically about 1% of the datasets required by AI-assisted EM emulators) is needed to construct this matrix. Subsequently, any arbitrary pixel configuration can be evaluated analytically using a closed-form multiport relation without additional full-wave calculations. The proposed approach eliminates data-driven overfitting and ensures accurate results across all design variations. Using MAPES, comprehensive examples for single- and double-layer PCBs and CMOS processes (180 nm and 65 nm) confirm that high prediction accuracy with 600-2000$\times$ speed improvement is achieved compared to CST simulations. Owing to its efficiency, scalability, and reliability, MAPES provides a practical and versatile tool for AI-assisted MW circuit and RFIC design across diverse fabrication technologies.
arXiv:2606.30338v1 Announce Type: new Abstract: External evaluations are becoming increasingly central to the governance of AI systems. In practice, however, independent auditors often have limited access to deployed models and must rely on query-based interactions. Most existing fairness evaluation methods assume static datasets and fixed-sample statistical tests, making them poorly suited to real-world auditing scenarios in which evidence must be collected sequentially under query constraints. In this work, we formulate fairness auditing as a tolerance-aware sequential hypothesis-testing problem under limited model output access. We develop a sequential generalized likelihood-ratio framework that allows auditors to accumulate evidence from a finite audit pool and stop once sufficient support for compliance or violation has been obtained. The framework is instantiated for decision-based Statistical Parity and Equal Opportunity audits, and extended to score- and logit-based proxy audits when richer observables are available. Our results show that both the fairness metric and the level of model access significantly affect audit efficiency, and that the benefits of richer output information are not uniform across auditing settings. In particular, richer outputs can substantially reduce the number of queries required for some fairness metrics and operating regimes, while offering limited gains in near-threshold cases. This work provides a practical statistical framework for sequential fairness auditing under realistic deployment constraints.
arXiv:2606.30339v1 Announce Type: new Abstract: Aligning large language models (LLMs) with diverse user preferences is a critical yet challenging task. While post-training methods can adapt models to specific needs, they often require costly data curation and additional training. Test-time scaling (TTS) presents an efficient, training-free alternative, but its application has been largely limited to verifiable domains like mathematics and coding, where response correctness is easily judged. To extend TTS to preference alignment, we introduce a novel framework that models the task as a realignment problem, since the base model often fails to sufficiently align with the stated preference. Our key insight is to decompose the underlying reward function into two components: one related to the question and the other to preference information. This allows us to derive a REAlignment Reward (REAR) that selectively rescales the proportions of these two reward terms. We then show that REAR can be formulated as a linear combination of token-level policy log-probabilities, making it computationally efficient and easy to integrate with various TTS algorithms such as best-of-$N$ sampling and tree search. Experiments show that compared to other test-time baselines, REAR not only enables scalable test-time realignment for preference alignment tasks under diverse user requirements, but also generalizes to mathematical and visual tasks under appropriate preference settings.
arXiv:2606.30251v1 Announce Type: new Abstract: Agentic multimodal models perform diverse operations on an image via code and reason over the returned view, an effective paradigm for fine-grained visual question answering. However, code operations can be useful, redundant, or misleading. Outcome-only rewards cannot precisely distinguish these cases, and existing process rewards either fail to attribute final correctness to individual tool calls, or require an external judge model. To address this, we introduce Tool-Augmented Credit Optimization (TACO), a GRPO variant for code-tool agents built on two coupled advantage channels. The first, Differential Answer-Probe Reward (DAPR), is a self-supervised, judge-free tool-contribution advantage that credits each tool call by its own effect on answering correctly. Probe tokens inserted into the model's reasoning elicit its predictions with and without the tool, and the difference in outcome reward is taken as the call's value: positive for a useful call, negative for a misleading one, and zero for one that changes nothing. This reuses the existing answer checker with no auxiliary judge, and, being a difference rather than an absolute probe score, is naturally robust to probe-hacking. The second is the outcome advantage from the final answer, distributed by Outcome-Gated Advantage Routing (OGAR): a parameter-free rule that, conditioned on the call's outcome, delivers this credit only to the responsible segments, suppressing wasted tool calls without any cost term. We train TACO through a two-stage SFT+RL pipeline. Extensive experiments across perception, reasoning, and general multimodal benchmarks show that it yields consistent accuracy gains and learns to invoke its tools only when they help.