arXiv:2511.03603v2 Announce Type: replace Abstract: Industrial embedded systems are typically used to execute simple control algorithms due to their low computational resources. Despite these limitations, the implementation of advanced control techniques such as Model Predictive Control (MPC) has been explored by the control community in recent years, typically considering simple linear formulations or explicit ones to facilitate the online computation of the control input. These simplifications often lack features and properties that are desirable in real-world environments. This article presents an efficient implementation for embedded systems of MPC for tracking with artificial reference, solved via a recently developed structure-exploiting ADMM-based algorithm. This formulation is tailored to a wide range of applications by incorporating essential practical features at a small computational cost, including integration with an offset-free scheme, back-off parameters that enable constraint tightening, and soft constraints that preserve feasibility under disturbances or plant-model mismatch. This is accompanied with a framework for probabilistic performance validation of the closed-loop system over long-term operation. The applicability of the approach is illustrated on a Programmable Logic Controller (PLC), incorporated in a hardware-in-the-loop setup to control a nonlinear continuous stirred-tank reactor. The behavior of the closed-loop system is probabilistically validated with respect to constraint violations and the number of iterations required at each time step by the MPC optimization algorithm.
Science Journals
arXiv:2606.08186v1 Announce Type: new Abstract: Monopedal hopping robots are conceptually simple but highly dynamic and inherently unstable. Achieving robust 3D hopping is still difficult because ground reaction forces are available only during the short stance phase, while the robot is underactuated in flight. A key unresolved issue is how to improve flight-phase control authority. Propeller assistance provides a promising solution, but it requires careful coordination of leg-generated contact forces and propeller thrusts across stance and flight. This paper presents Pro-OMEGA2, a propeller-assisted 3D monopedal hopping robot with an active 3-RSR parallel leg and a trunk-mounted tri-rotor for auxiliary attitude regulation. To address the force coordination challenge, we propose a Hierarchical Force Allocation (HFA) framework based on a single rigid body (SRB) model. The leg generates the main stance contact wrench, while the tri-rotor provides auxiliary attitude regulation, compensating the residual attitude moment in stance and maintaining attitude during flight. Real-robot experiments in indoor and outdoor scenarios demonstrate sustained 3D hopping, including terrain transitions and impulsive push recovery, validating robustness under unmodeled contact and external disturbances.
arXiv:2511.03877v2 Announce Type: replace Abstract: Social and collaborative platforms emit multivariate time-series traces in which early interactions -- such as views, likes, or downloads -- are followed, sometimes months or years later, by higher impact like citations, sales, or reviews. We formalize this setting as Lead-Lag Forecasting (LLF): given an early usage channel (the lead), predict a correlated but temporally shifted outcome channel (the lag). Despite the ubiquity of such patterns, LLF has not been treated as a unified forecasting problem within the time-series community, largely due to the absence of standardised datasets. To anchor research in LLF, here we present two high-volume benchmark datasets: arXiv (accesses -> citations of 2.3M papers) and GitHub (pushes/stars -> forks of 3M repositories). Our datasets provide ideal testbeds for lead-lag forecasting, by capturing long-horizon dynamics across years, spanning the full spectrum of outcomes, and avoiding survivorship bias in sampling. We documented all technical details of data curation and cleaning, verified the presence of lead-lag dynamics through statistical and classification tests, and benchmarked parametric and non-parametric baselines for regression. Our study establishes LLF as a novel forecasting paradigm and lays an empirical foundation for its systematic exploration in social and usage data.
arXiv:2511.04124v3 Announce Type: replace Abstract: Symbolic regression (SR) models complex systems by discovering mathematical expressions that capture underlying relationships in observed data. However, most SR methods prioritize minimizing prediction error over identifying the governing equations, often producing overly complex or inaccurate expressions. To address this, we present a decomposable SR method that generates interpretable multivariate expressions leveraging transformer models, genetic algorithms (GAs), and genetic programming (GP). In particular, our explainable SR method distills a trained ``opaque'' regression model into mathematical expressions that serve as explanations of its computed function. Our method employs a Multi-Set Transformer to generate multiple univariate symbolic skeletons that characterize how each variable influences the opaque model's response. We then evaluate the generated skeletons' performance using a GA-based approach to select a subset of high-quality candidates before incrementally merging them via a GP-based cascade procedure that preserves their original skeleton structure. The final multivariate skeletons undergo coefficient optimization via a GA. We evaluated our method on problems with controlled and varying degrees of noise, demonstrating lower or comparable interpolation and extrapolation errors compared to two GP-based methods, three neural SR methods, and a hybrid approach. Unlike them, our approach consistently learned expressions that matched the original mathematical structure. Similarly, our method achieved both a high symbolic solution recovery rate and competitive predictive performance relative to benchmark methods on the Feynman dataset.
arXiv:2604.02056v2 Announce Type: replace Abstract: Missing modalities in multimodal sensing cause not only information loss but also a fusion-interface mismatch: a fusion head trained on a canonical set of modality slots must operate on changing observed subsets at inference time. We propose Compass, an interface-complete fusion framework that restores this canonical slot structure before prediction. Each modality is assigned a fixed fusion slot. Observed modalities populate their slots with real representations, while absent modalities are filled with target-slot completion representations estimated from the observed sources. Multiple source-specific estimates for the same missing slot are aggregated into a single slot filler, allowing the same lightweight fusion operator to be applied under arbitrary missing-modality patterns. Training uses synthetic modality masking, slot-compatibility supervision, and representation-space stabilization to make completed slots compatible with real modality representations and useful for downstream recognition. Across XRF55, MM-Fi, and OctoNet, Compass improves robustness under diverse single- and multiple-missing settings, including controlled comparisons against imputation, distillation, and translation-style baselines. These results suggest that preserving the fusion interface is a simple and effective principle for robust multimodal sensing.
arXiv:2604.07848v2 Announce Type: replace Abstract: Multi-task learning shows strikingly inconsistent results -- sometimes joint training helps substantially, sometimes it actively harms performance -- yet the field lacks a principled framework for predicting these outcomes. We identify a fundamental but unstated assumption underlying gradient-based task analysis: tasks must share training instances for gradient conflicts to reveal genuine relationships. When tasks are measured on the same inputs, gradient alignment reflects shared mechanistic structure; when measured on disjoint inputs, any apparent signal conflates task relationships with distributional shift. We discover this sample overlap requirement exhibits a sharp phase transition: below 30% overlap, gradient-task correlations are statistically indistinguishable from noise; above 40%, they reliably recover known biological structure. Comprehensive validation across multiple datasets achieves strong correlations and recovers biological pathway organization. Standard benchmarks systematically violate this requirement -- MoleculeNet operates at <5% overlap, TDC at 8-14% -- far below the threshold where gradient analysis becomes meaningful. This provides the first principled explanation for seven years of inconsistent MTL results.
arXiv:2606.09501v1 Announce Type: new Abstract: Understanding quantum dynamics in curved spacetime is a central challenge at the intersection of quantum mechanics and gravity. Anti-de-Sitter (AdS) spacetime plays a pivotal role in the context of the AdS/CFT correspondence, which relates gravitational dynamics in the AdS bulk to a conformal field theory (CFT) living on its boundary. Despite its foundational importance, direct experimental access to dynamical quantum phenomena in Lorentzian AdS spacetime has so far remained out of reach. Here, we report the first experimental emulation of fermionic wave packet dynamics in Lorentzian AdS spacetime using a photonic platform. By mapping the Dirac equation in curved spacetime onto the propagation of light in engineered wave\-guide arrays, we directly observe gravitational confinement of relativistic wave packets and resolve their center-of-mass motion in real time. We identify a characteristic superposition of slow geodesic oscillations governed solely by spacetime curvature and fast Zitterbewegung arising from relativistic particle--antiparticle interference. While the geodesic frequency is independent of fermion mass, the Zitterbewegung frequency exhibits a distinct joint dependence on mass and curvature, revealing a curvature-induced modification of relativistic quantum dynamics. Our results provide the first quantitative experimental access to fermionic bulk dynamics in emulated AdS$_2$ spacetime with Lorentzian signature. This establishes a scalable analog platform that may potentially be used for exploring dynamical aspects of holography.
arXiv:2606.09502v1 Announce Type: new Abstract: This paper is devoted to the numerical analysis of a second-order bulk--surface splitting scheme for the semi-linear wave equation with kinetic boundary conditions. The construction is based on the interpretation of the equations as coupled system and the implementation of different difference formulae for the discrete states depending on their exact position in the system equations. This results in a 4-step scheme which decouples bulk and surface dynamics. We prove energy stability and second-order convergence under a weak CFL condition and validate these results also numerically.
arXiv:2606.09505v1 Announce Type: new Abstract: The split covariance intersection filter (Split CIF) is a useful tool for general data fusion and has the potential to be applied in a variety of engineering tasks. The w-optimization problem involved in the Split CIF concerns the performance and implementation efficiency of the Split CIF. It is known that the w-optimization problem enjoys the desirable property of convexity (or more clearly, the second-order convexity in this paper's context). This paper proves that the w-optimization problem further enjoys a more desirable property namely the fourth-order convexity, thanks to which a guaranteed fast implementation of the Split CIF can be realized. The new implementation is coined as the nested Newton method, which is also presented in this paper.
arXiv:2509.18712v3 Announce Type: replace Abstract: Optimality of several quasi-Monte Carlo methods and suboptimality of the sparse-grid quadrature based on the univariate Gauss--Hermite rule is proved in the Sobolev spaces of mixed dominating smoothness of order $\alpha$, where the optimality is in the sense of worst-case convergence rate. For sparse-grid Gauss--Hermite quadrature, lower and upper bounds are established, with rates coinciding up to a logarithmic factor. The dominant rate is found to be only $N^{-\alpha/2}$ with $N$ function evaluations, although the optimal rate is known to be $N^{-\alpha}(\ln N)^{(d-1)/2}$. The lower bound is obtained by exploiting the structure of the Gauss--Hermite nodes and is independent of the quadrature weights; consequently, no modification of the weights can improve the rate $N^{-\alpha/2}$. In contrast, several quasi-Monte Carlo methods with a change of variables are shown to achieve the optimal rate, some up to, and one including, the logarithmic factor.
arXiv:2606.09507v1 Announce Type: new Abstract: Video world models have made rapid progress in generating controllable visual experiences, but most of them still simulate the world from a single observer. Extending such models to multiple agents raises a central challenge: if each agent's future state is generated independently, overlapping views may instantiate different versions of the same scene, leading to inconsistent objects, layouts, and appearances across agents. Conventional camera conditioning controls individual trajectories, but it does not explicitly couple the generation of views that should agree under shared scene geometry. We introduce Prisma-World, a camera-controllable multi-agent world model that formulates multi-agent generation as a joint geometry-aware denoising process for cross-view consistency. Prisma-World processes all agent videos within one full-attention sequence, uses a multi-agent RoPE design to distinguish agent identities while preserving synchronized temporal coordinates, and injects relative camera geometry into attention to bias overlapping viewpoints toward shared scene evidence. To further strengthen multi-view consistency and enhance global spatial perception, we augment our framework with an overlap-decaying curriculum training paradigm alongside minimap-conditioned structural guidance. To facilitate the training and evaluation of multi-agent models, we introduce PrismaDataset, a large-scale UE5 dataset with panoramic acquisition across diverse scenes, composable multi-agent view groups with flexible agent counts and complex camera trajectories, and precise camera/action annotations for consistency training and evaluation. Experiments show that a single Prisma-World model can generate high-fidelity multi-agent videos with flexible agent numbers, camera controllability, improved cross-view consistency, and spatial grounding under minimap guidance.
arXiv:2606.09508v1 Announce Type: new Abstract: Existing sparse attention and KV cache compression methods for long-context LLM inference typically apply fixed sparsity patterns or uniform budgets across all attention heads, overlooking the substantial variation in attention behavior among heads and contexts. We observe two distinct entropy patterns among attention heads: Rigid Heads, whose entropy stays near zero across input segments, and Dynamic Heads, whose entropy fluctuates significantly. Crucially, the distribution of these types is context-dependent and cannot be predetermined offline. We therefore propose EntropyInfer, a training-free framework that uses attention entropy to adaptively allocate compute at the granularity of individual heads and segments during prefilling. For decoding, we introduce a latent KV cache compression scheme that leverages generated output tokens, rather than prefill tokens alone, to identify and retain the most critical cache entries. Extensive experiments on Llama, Qwen and openPangu model series show that EntropyInfer consistently outperforms baselines including SnapKV, AdaKV, and CritiPrefill, achieving up to 2.39$\times$ end-to-end speedup beyond 100k tokens with minimal quality degradation compared to full attention. The code is released in https://github.com/SHA-4096/EntropyInfer.
How Context Shapes Truth: Geometric Transformations of Statement-level Truth Representations in LLMs
arXiv:2601.06599v2 Announce Type: replace Abstract: Large Language Models (LLMs) often encode whether a statement is true as a vector in their residual stream activations. These vectors, also known as truth vectors, have been studied in prior work, however how they change when context is introduced remains unexplored. We study this question by measuring (1) the directional change ($\theta$) between the truth vectors with and without context and (2) the relative magnitude of the truth vectors upon adding context. Across four LLMs and four datasets, we find that (1) truth vectors are roughly orthogonal in early layers, converge in middle layers, and may stabilize or continue increasing in later layers; (2) adding context generally increases the truth vector magnitude, i.e., the separation between true and false representations in the activation space is amplified; (3) larger models distinguish relevant from irrelevant context mainly through directional change ($\theta$), while smaller models show this distinction through magnitude differences. We also find that context conflicting with parametric knowledge produces larger geometric changes than parametrically aligned context. To the best of our knowledge, this is the first work that provides a geometric characterization of how context transforms the truth vector in the activation space of LLMs.
arXiv:2601.06649v2 Announce Type: replace Abstract: Research in machine learning has questioned whether increases in training token counts reliably produce proportional performance gains in large language models. Building on prior work introducing an energy-aware parameter efficiency metric, this study empirically examines the effects of increasing training token counts under fixed hardware and training conditions. The significance of this work lies in the explicit integration of power consumption and execution duration, as reflected by the power sampling frequency, into token-scale analysis. This addresses a gap in prior studies emphasizing performance outcomes while underrepresenting computational and energy costs. Using a repeated-measures experimental design on a constant GPU instance with an identical model architecture, optimizer settings, and epoch counts, a 1.1-billion-parameter TinyLlama model was trained at three token counts (500K, 1M, and 2M). While conventional performance metrics exhibited inconsistent or diminishing returns across token scales, the inclusion of power consumption and execution duration revealed a strictly monotonic decline in training efficiency as token count increased. Repeated-measures ANOVA demonstrated a strong effect of token count on parameter efficiency, with all pairwise comparisons remaining significant following Bonferroni correction. These findings indicate that increases in training token counts may be energetically inefficient even when marginal performance improvements are observed, underscoring the importance of efficiency-aware evaluation in large language model training.
arXiv:2606.08200v1 Announce Type: new Abstract: Evaluating LLM-powered interactive social agents is challenging because socially relevant behaviors depend not only on isolated outputs, but also on prior interactions, social roles, and downstream actions. Existing methods typically allow a target agent to act freely in an environment and then score the resulting trajectory. However, this passive setup can miss capabilities that only become observable under specific social circumstances; for example, conflict handling may remain untested if no disagreement arises. We propose Online Agent-as-a-Judge, a situation-generating evaluation framework for interactive social agents. Online Agent-as-a-Judge deploys an in-world evaluator agent that interacts with the target agent through the environment's native dialogue and action protocol, actively eliciting situations relevant to the evaluation criteria. The resulting trajectories provide evidence for assessing both immediate responses and subsequent behavior. In a life-simulation environment with $32$ designer-authored social criteria, Online Agent-as-a-Judge improves criteria coverage and agreement with human labels, yielding more reliable evidence-grounded evaluations of behaviors that passive methods can leave unobserved.
arXiv:2606.09130v1 Announce Type: new Abstract: Singlet fission converts an optically excited singlet state into a spin-entangled triplet pair state (TT$_1$)$^1$ that can, in principle, yield two free triplets for photovoltaics and/or a polarized high spin state for quantum technologies. Synthetically tunable templates suggest that the above photophysics is governed by a subtle but poorly understood interplay of molecular motifs, geometry and structural fluctuations. Here, we investigate the (TT$_1$)$^1$ state in a library of conformationally flexible pentacenic dimers, where a (TT$_1$)$^1$-specific near-IR spectral feature is readily available. Using a suite of polarization-controlled impulsive optical spectroscopies, we find that (TT$_1$)$^1$ formation is specific to planar conformations and is accompanied by large nuclear reorganization in the (TT$_1$)$^1$ photoproduct. Introducing polarization anisotropy to track the electronic character of the (TT$_1$)$^1$ species, supported by screened configuration interaction based electronic structure theory, we find that significant singlet-triplet electronic mixing is persistent throughout its evolution. This behavior is universal across diverse bridging motifs and indicates that, once the triplet pair is strongly bound, neither substantial nuclear reorganization nor structural fluctuations on longer timescales are sufficient to suppress persistent singlet-triplet electronic mixing, such that triplet-pair decorrelation is outcompeted by its decay. Our observations establish polarization-selective pump-probe and anisotropy as a direct optical probe of triplet pair decorrelation, complementary to spin-selective measurements at longer timescales.
arXiv:2604.17324v2 Announce Type: replace Abstract: Global self-attention drives modern graph transformers, yet the softmax at its core imposes a structural constraint rarely examined directly: every attention row is non-negative and sums to one, so each per-head output is a mass-conserving convex combination of value vectors. A node can never "attend to nothing." We argue this conservation constraint is a single root cause behind three pathologies usually studied in isolation: the collapse of node representations with depth (over-smoothing), a low-rank bottleneck on per-head outputs, and brittle optimization in deep stacks. Drawing on how sigmoid gating removes analogous attention sinks in language models, we introduce SigGate-GT, a graph transformer that applies a learned, per-head, input-conditioned sigmoid gate to the attention output inside the GraphGPS framework. The gate is a smooth, per-dimension "volume control" that can drive head outputs toward zero, relaxing the constraint without abandoning attention's probabilistic interpretation. Analytically and through synthetic experiments, we show the gate strictly increases the stable rank of per-head outputs, and connect this rank gain to all three manifestations. On five molecular and long-range benchmarks, SigGate-GT matches the prior best on ZINC (0.059 MAE), records the strongest result among the graph-transformer baselines we evaluate on ogbg-molhiv (82.47% ROC-AUC), and is competitive on ogbg-molpcba and the Long-Range Graph Benchmark, with statistically significant gains over GraphGPS on all five datasets (p < 0.05). Mechanism analyses confirm the diagnosis: gating slows over-smoothing (a 30% mean relative gain in representation diversity across 4-16 layers), keeps attention entropy from collapsing, and stabilizes training across a 10x learning-rate range, at about 1% parameter overhead on OGB and under 3% wall-clock cost.
arXiv:2606.09553v1 Announce Type: new Abstract: Recent advances in neural text-to-speech (TTS) and multilingual speech generation have substantially improved synthetic speech quality, yet these gains remain unevenly distributed across the world's languages. Existing models are still dominated by a small set of high-resource languages, while many studies of low-resource TTS are simulated on artificially downsampled high-resource corpora that do not reflect the orthographic variation and limited phonetic coverage encountered in genuinely underrepresented settings. As such, we introduce OpenBibleTTS, which is a large-scale benchmark for low-resource speech synthesis spanning 37 underrepresented languages. Moreover, a systematic comparison of various TTS architectures and large-scale speech generation models is conducted across in-domain Biblical text and out-of-domain material. Results show that no single system dominates across languages and metrics: Gemini-TTS achieves the highest listener ratings on most evaluated languages, but monolingual EveryVoice models trained on OpenBibleTTS remain strongest for intelligibility and are preferred in several African languages, while open from-scratch systems degrade sharply on out-of-domain text, revealing a persistent gap between broad multilingual coverage and reliable synthesis quality in underserved linguistic communities. We complement automatic evaluation with subjective human judgments, and open-source all processed datasets, alignments, and trained models to support future low-resource TTS research.
arXiv:2606.09554v1 Announce Type: new Abstract: Optical trapping and micromanipulation rely on carefully shaped electromagnetic fields to exert precise forces and torques on microscopic particles. Despite their widespread application in biology and nanotechnology, the absolute physical limits of trapping performance, specifically the maximum achievable optical force and trap stiffness, have not yet been rigorously quantified. This work establishes a general theoretical framework to determine these fundamental bounds in the dipole approximation. By relating the optical force and stiffness to a local Taylor expansion of the electromagnetic field at the particle location, we formulate the performance limit as a solution to a quadratically constrained quadratic program. To evaluate these bounds, we employ two complementary approaches. First, we utilize a complete basis of vector spherical wave functions to determine the absolute theoretical limits of optical force and stiffness permitted by Maxwell's equations in free space, revealing Pareto-optimal trade-offs between stable confinement and directional force. Second, we introduce an aperture-based formulation that restricts the incident fields to those realizable by finite planar apertures. This yields device-consistent bounds directly applicable to experimental setups which rely mostly on electromagnetic beams. The finding that optimized aperture fields can outperform standard Gaussian beams by removing the severe axial bottleneck is particularly important. By comparing these two regimes, we identify the specific spatial modes that contribute to stable trapping and quantify the performance trade-offs inherent to physical beam shaping. This dual framework provides provably optimal bounds for power-normalized optical tweezers and serves as a rigorous benchmark for evaluating realistic beam designs.
arXiv:2606.08238v1 Announce Type: new Abstract: Constitutive modeling of the relationship between process-imposed material states and fundamental material properties is critical to control of material microstructure in manufacturing processes. The limited accuracy resulting from the typical reliance on fallible human expertise and intuition for postulation and revision of the models functional form results in incremental and time consuming model discovery. Conventional Machine Learning (ML) incurs significant cost and time of data generation. Model discovery using Large Language Models (LLMs) suffers from the above issues and/or ignores the inviolability of fundamental thermodynamics laws. This work creates a novel GPT-Micro paradigm for autonomous, data sparse, and thermodynamics-compliant discovery of de-novo constitutive models. This framework seamlessly integrates semantic knowledge extraction from literature, enforcement of thermodynamics-based conservation laws, and sparse datasets, with LLM-driven generation and refinement of model hypotheses. Validation is performed for a long-intractable constitutive modeling problem in a printed electronics process testbed. This reveals significant and simultaneous advantages over the state-of-the-art including: (a) More than 70 percent reduction in data burden relative to ML-based modeling without loss in accuracy; (b) 400X reduction in discovery time after data generation, from months to hours, relative to human-driven modeling; (c) Discovery of models with novel functional forms without subjective human choice of a starting hypothesis; (d) Enhanced physics-rooted trustworthiness, human interpretability, and mechanistic insight via synthesis of compact, conservation-compliant, and physically complete analytical models. The potential of GPT-Micro to realize rapid, low-cost, physically trustworthy, and interpretable microstructure modeling across the manufacturing landscape is discussed.
arXiv:2606.08239v1 Announce Type: new Abstract: Multimodal large language models (MLLMs) have made substantial advancements in video understanding, yet the reliability of their responses remains underexplored. This work presents a diagnostic study of absent answer detection for MLLMs in video understanding, where the correct answer is deliberately excluded from the candidate set and a reliable model is expected to recognize that no valid option exists. We evaluate the absent answer detection behavior under three settings: multiple-choice questions augmented with an ``None of the Above'' option, open-ended generation with a detection instruction, and standard evaluation without any guidance. Across a diverse set of models and benchmarks, we find that MLLMs overwhelmingly select plausible distractors rather than detecting the absent answer. This failure is more pronounced in temporal reasoning tasks and worsens with denser frame sampling. We further explore chain-of-thought prompting as a mitigation strategy and find that while it substantially improves detection rates, performance remains unsatisfactory, suggesting that prompting-based strategies alone are insufficient to fully address this limitation. These findings expose a systematic failure in absent answer detection and highlight the need for explicit detection mechanisms in multimodal systems.
arXiv:2603.25157v3 Announce Type: replace Abstract: Recent vision backbones, such as Transformer families and state-space models like Mamba, have achieved remarkable progress on image recognition. Despite their empirical success, these architectures remain far from the computational principles of the human brain, often demanding enormous amounts of training data while offering limited interpretability. We propose the Vision Hopfield Memory Network (V-HMN), a brain-inspired vision backbone that integrates hierarchical memory mechanisms across layers with iterative refinement updates. Specifically, V-HMN incorporates local Hopfield modules that provide associative memory dynamics at the image patch level, global Hopfield modules that function as episodic memory for contextual modulation, and a predictive-coding-inspired refinement rule for iterative error correction. By organizing these memory-based modules hierarchically, V-HMN captures both local and global dynamics in a unified framework. Memory retrieval exposes the relationship between inputs and stored patterns, providing a prototype-based form of interpretability through explicit memory retrieval, while the reuse of stored patterns improves data efficiency. This brain-inspired design therefore enhances data efficiency and provides a prototype-based form of interpretability compared to existing self-attention- or state-space-based approaches. We conducted extensive experiments on public image classification benchmarks. V-HMN achieves strong performance on small- and medium-scale benchmarks, and remains competitive with widely adopted backbone architectures on ImageNet despite minimal architectural tuning, while offering improved data efficiency and a prototype-based form of interpretability. These findings highlight the potential of V-HMN as a memory-centric alternative to standard vision backbones, thereby bridging brain-inspired computation with modern machine learning.
arXiv:2604.02272v2 Announce Type: replace Abstract: We present an X-ray characterization of a fully depleted, 725 $\mu$m thick p-channel SiSeRO-CCD. Measurements with a $^{55}$Fe source yield an energy resolution of $54 \pm 0.9$ eV ($14.6 \pm 0.25 e^{-}$) at 5.9 keV for single-pixel events, indicating that the SiSeRO amplifier preserves the intrinsic charge resolution of the CCD under multi-sample non-destructive readout. Characterization with a $^{241}$Am source extends the response to higher-energy photons, with reconstructed spectral features observed between 9-26 keV and the 59.5 keV $\gamma$ emission. These measurements, together with a muon-derived diffusion calibration, show that charge transport and diffusion are consistent with interactions spanning the full sensor depth. These results demonstrate that the SiSeRO-CCD simultaneously achieves sub-electron noise performance and efficient charge collection in a thick, fully depleted silicon detector. This combination enables X-ray spectroscopy across a broad energy range while maintaining sensitivity to faint signals.
arXiv:2606.09556v1 Announce Type: new Abstract: AI Scientist agents are often evaluated as if capability were mainly a function of model quality, prompting, or reasoning scaffolds. We test a different hypothesis in drug-asset valuation: for knowledge-intensive scientific decisions, the limiting factor is often the evidence substrate the agent can access. We run a controlled three-arm ablation on a production valuation agent: A is a plain web-only LLM analyst, B adds public structured tools plus a 14-dimension valuation playbook, verifier, objectivity policy and red-team, and C adds the proprietary Noah AI corpus of curated pipeline, trial and deal intelligence. Across a 13-asset stratified benchmark, B improves calibration and audit discipline: tier-in-range accuracy rises from 0.80 to 0.89 and objectivity from 3.16 to 3.30. But B does not remove the factual ceiling. Under capability-superset accounting, A and B recover only 0.25 and 0.38 of the curated gold competitive record, while C recovers 0.96; on the curated long-tail subset, C reaches 0.93 vs. 0.26/0.30. Raw blind-panel decision quality is similar for A and B (7.01 vs. 6.96), so we introduce completeness-aware decision utility: informed decision-quality = decision-quality x gold-coverage. On this metric, C reaches 7.43 vs. 1.76/2.57 for A/B. Even a perfect non-proprietary-data report would be capped at 3.83 by B's coverage. The result is not that reasoning scaffolds are unimportant; they improve calibration and discipline. Rather, proprietary evidence sets the upper bound of what the AI Scientist can know and therefore decide.
arXiv:2606.09565v1 Announce Type: new Abstract: Nonconcave games present a unique challenge, as neither pure Nash equilibria nor local Nash equilibria (LNE) are guaranteed to exist, even in zero-sum settings. Additionally, computing approximate LNE in smooth multiplayer games over bounded regions is PPAD-hard. These challenges, coupled with the inherent complexity, have driven recent research toward broader equilibrium concepts, such as min-max critical points, and first-order Nash equilibria (FONE), which correspond to solutions of specific non-monotone variational inequalities. This paper addresses general-sum multiplayer games with compact convex strategy sets and smooth, nonconcave utility functions. Daskalakis et al. introduced the STON'R algorithm for solving variational inequality problems and established convergence under smoothness assumptions. They further showed that the algorithm's limit points correspond to equilibria in specific classes of games, namely local minimax equilibria in two-player zero-sum games and Nash equilibria in smooth concave games. In this work, we extend the convergence result to multiplayer general-sum games and show that the variational inequality solutions targeted by STON'R correspond to first-order Nash equilibria (FONE), a general game-theoretic solution concept that unifies these previously studied cases. We demonstrate the effectiveness and robustness of the algorithm on various examples from recent literature.