arXiv:2511.09527v1 Announce Type: new 
Abstract: Machine learning fits model parameters to approximate input-output mappings, predicting unknown samples. However, these models often require extensive arithmetic computations during inference, increasing latency and power consumption. This paper proposes a digital-time-domain computing approach for Tsetlin machine (TM) inference process to address these challenges. This approach leverages a delay accumulation mechanism to mitigate the costly arithmetic sums of classes and employs a Winner-Takes-All scheme to replace conventional magnitude comparators. Specifically, a Hamming distance-driven time-domain scheme is implemented for multi-class TMs. Furthermore, differential delay paths, combined with a leading-ones-detector logarithmic delay compression digital-time-domain scheme, are utilised for the coalesced TMs, accommodating both binary-signed and exponential-scale delay accumulation issues. Compared to the functionally equivalent, post-implementation digital TM architecture baseline, the proposed architecture demonstrates orders-of-magnitude improvements in energy efficiency and throughput.

تقدم ورقة جديدة نهجًا مبتكرًا للحوسبة الرقمية في مجال الزمن لتفسير آلات تسيتلين، مما يعالج مشكلات الكمون العالي واستهلاك الطاقة في نماذج التعلم الآلي. من خلال استخدام آلية تراكم التأخير ونظام 'Winner-Takes-All'، تحسن هذه الطريقة بشكل كبير من كفاءة الطاقة ومعدل الإنتاج، مما يمثل تقدمًا كبيرًا في مجال الذكاء الاصطناعي.

Un nuevo artículo presenta un enfoque innovador de computación digital en el dominio del tiempo para la inferencia de máquinas de Tsetlin, abordando problemas de alta latencia y consumo de energía en modelos de aprendizaje automático. Al utilizar un mecanismo de acumulación de retrasos y un esquema de 'Winner-Takes-All', este método mejora significativamente la eficiencia energética y el rendimiento, marcando un avance sustancial en el campo de la inteligencia artificial.

Un nouvel article présente une approche innovante de calcul numérique en temps réel pour l'inférence des machines de Tsetlin, abordant les problèmes de latence élevée et de consommation d'énergie dans les modèles d'apprentissage automatique. En utilisant un mécanisme d'accumulation de délais et un schéma 'Winner-Takes-All', cette méthode améliore considérablement l'efficacité énergétique et le débit, marquant une avancée substantielle dans le domaine de l'intelligence artificielle.

A new paper presents an innovative digital-time-domain computing approach for Tsetlin machine inference, addressing high latency and power consumption issues in machine learning models. By utilizing a delay accumulation mechanism and a Winner-Takes-All scheme, this method significantly enhances energy efficiency and throughput, marking a substantial advancement in the field of artificial intelligence.

Event-Driven Digital-Time-Domain Inference Architectures for Tsetlin Machines

arXiv:1906.07172v5 Announce Type: replace-cross 
Abstract: Equivariant neural networks are a class of neural networks designed to preserve symmetries inherent in the data. In this paper, we introduce a general method for modifying a neural network to enforce equivariance, a process we refer to as equivarification. We further show that group convolutional neural networks (G-CNNs) arise as a special case of our framework.

Equivariant neural networks and equivarification

arXiv:2311.13745v4 Announce Type: replace-cross 
Abstract: Diffusion models have become the most popular approach to deep generative modeling of images, largely due to their empirical performance and reliability. From a theoretical standpoint, a number of recent works have studied the iteration complexity of sampling, assuming access to an accurate diffusion model. In this work, we focus on understanding the sample complexity of training such a model; how many samples are needed to learn an accurate diffusion model using a sufficiently expressive neural network? Prior work showed bounds polynomial in the dimension, desired Total Variation error, and Wasserstein error. We show an exponential improvement in the dependence on Wasserstein error and depth, along with improved dependencies on other relevant parameters.

Improved Sample Complexity Bounds for Diffusion Model Training

arXiv:2410.22995v2 Announce Type: replace-cross 
Abstract: A hallmark of advanced artificial intelligence is the capacity to progress from passive visual perception to the strategic modification of visual information to facilitate complex reasoning. This advanced capability, however, remains critically underdeveloped in current Large Multi-modal Models (LMMs). The deficiency is often masked by evaluation metrics that prioritize final-answer accuracy, creating an illusion of competence where genuine reasoning is absent. Using the domain of geometric problem-solving as a precise instrument, we probe this issue through tasks that require constructing visual aids. To this end, we introduce \textbf{VisAidMath}, a challenging benchmark, and our novel Three-Layered Funnel Evaluation Framework. This framework moves beyond simple accuracy (ACCU) to scrutinize the generation of valid visual aids (PVA) and the soundness of subsequent reasoning steps (SPRS). Our extensive experiments on state-of-the-art models, including Doubao-Seed-1.6 and o4, reveal a profound ``Reasoning Illusion''. We observe that high surface-level accuracy conceals a catastrophic failure in the models' ability to produce valid visual aids or to reason from them. Our findings expose a fundamental schism between visual perception and logical deduction in modern LMMs. We host an evaluation platform at CodaBench for testing publicly. Homepage: https://nlp2ct.github.io/VisAidMathHomepage/ Evaluation: https://www.codabench.org/competitions/7634/

VisAidMath: Benchmarking Visual-Aided Mathematical Reasoning

arXiv:2508.01249v3 Announce Type: replace-cross 
Abstract: Large Language Model (LLM) agents offer a powerful new paradigm for solving various problems by combining natural language reasoning with the execution of external tools. However, their dynamic and non-transparent behavior introduces critical security risks, particularly in the presence of prompt injection attacks. In this work, we propose a novel insight that treats the agent runtime traces as structured programs with analyzable semantics. Thus, we present AgentArmor, a program analysis framework that converts agent traces into graph intermediate representation-based structured program dependency representations (e.g., CFG, DFG, and PDG) and enforces security policies via a type system. AgentArmor consists of three key components: (1) a graph constructor that reconstructs the agent's runtime traces as graph-based intermediate representations with control and data flow described within; (2) a property registry that attaches security-relevant metadata of interacted tools \& data, and (3) a type system that performs static inference and checking over the intermediate representation. By representing agent behavior as structured programs, AgentArmor enables program analysis for sensitive data flow, trust boundaries, and policy violations. We evaluate AgentArmor on the AgentDojo benchmark, the results show that AgentArmor can reduce the ASR to 3\%, with the utility drop only 1\%.

AgentArmor: Enforcing Program Analysis on Agent Runtime Trace to Defend Against Prompt Injection

arXiv:2511.14691v1 Announce Type: cross 
Abstract: Attention is the brain's ability to selectively focus on a few specific aspects while ignoring irrelevant ones. This biological principle inspired the attention mechanism in modern Transformers. Transformers now underpin large language models (LLMs) such as GPT, but at the cost of massive training and inference energy, leading to a large carbon footprint. While brain attention emerges from neural circuits, Transformer attention relies on dot-product similarity to weight elements in the input sequence. Neuromorphic computing, especially spiking neural networks (SNNs), offers a brain-inspired path to energy-efficient intelligence. Despite recent work on attention-based spiking Transformers, the core attention layer remains non-neuromorphic. Current spiking attention (i) relies on dot-product or element-wise similarity suited to floating-point operations, not event-driven spikes; (ii) keeps attention matrices that suffer from the von Neumann bottleneck, limiting in-memory computing; and (iii) still diverges from brain-like computation. To address these issues, we propose the Spiking STDP Transformer (S$^{2}$TDPT), a neuromorphic Transformer that implements self-attention through spike-timing-dependent plasticity (STDP), embedding query--key correlations in synaptic weights. STDP, a core mechanism of memory and learning in the brain and widely studied in neuromorphic devices, naturally enables in-memory computing and supports non-von Neumann hardware. On CIFAR-10 and CIFAR-100, our model achieves 94.35\% and 78.08\% accuracy with only four timesteps and 0.49 mJ on CIFAR-100, an 88.47\% energy reduction compared to a standard ANN Transformer. Grad-CAM shows that the model attends to semantically relevant regions, enhancing interpretability. Overall, S$^{2}$TDPT illustrates how biologically inspired attention can yield energy-efficient, hardware-friendly, and explainable neuromorphic models.

Attention via Synaptic Plasticity is All You Need: A Biologically Inspired Spiking Neuromorphic Transformer

arXiv:2511.08955v2 Announce Type: replace-cross 
Abstract: Simulating microstructure evolution (MicroEvo) is vital for materials design but demands high numerical accuracy, efficiency, and physical fidelity. Although recent studies on deep learning (DL) offer a promising alternative to traditional solvers, the field lacks standardized benchmarks. Existing studies are flawed due to a lack of comparing specialized MicroEvo DL models with state-of-the-art spatio-temporal architectures, an overemphasis on numerical accuracy over physical fidelity, and a failure to analyze error propagation over time. To address these gaps, we introduce MicroEvoEval, the first comprehensive benchmark for image-based microstructure evolution prediction. We evaluate 14 models, encompassing both domain-specific and general-purpose architectures, across four representative MicroEvo tasks with datasets specifically structured for both short- and long-term assessment. Our multi-faceted evaluation framework goes beyond numerical accuracy and computational cost, incorporating a curated set of structure-preserving metrics to assess physical fidelity. Our extensive evaluations yield several key insights. Notably, we find that modern architectures (e.g., VMamba), not only achieve superior long-term stability and physical fidelity but also operate with an order-of-magnitude greater computational efficiency. The results highlight the necessity of holistic evaluation and identify these modern architectures as a highly promising direction for developing efficient and reliable surrogate models in data-driven materials science.

MicroEvoEval: A Systematic Evaluation Framework for Image-Based Microstructure Evolution Prediction

arXiv:2511.11498v1 Announce Type: cross 
Abstract: We consider the problems of \emph{learning} and \emph{testing} real-valued convex functions over Gaussian space. Despite the extensive study of function convexity across mathematics, statistics, and computer science, its learnability and testability have largely been examined only in discrete or restricted settings -- typically with respect to the Hamming distance, which is ill-suited for real-valued functions.
  In contrast, we study these problems in high dimensions under the standard Gaussian measure, assuming sample access to the function and a mild smoothness condition, namely Lipschitzness. A smoothness assumption is natural and, in fact, necessary even in one dimension: without it, convexity cannot be inferred from finitely many samples. As our main results, we give:
  - Learning Convex Functions: An agnostic proper learning algorithm for Lipschitz convex functions that achieves error $\varepsilon$ using $n^{O(1/\varepsilon^2)}$ samples, together with a complementary lower bound of $n^{\mathrm{poly}(1/\varepsilon)}$ samples in the \emph{correlational statistical query (CSQ)} model.
  - Testing Convex Functions: A tolerant (two-sided) tester for convexity of Lipschitz functions with the same sample complexity (as a corollary of our learning result), and a one-sided tester (which never rejects convex functions) using $O(\sqrt{n}/\varepsilon)^n$ samples.

يتناول المقال تعلم واختبار الدوال المحدبة ذات القيم الحقيقية في الفضاء الغاوسي. يسلط الضوء على قيود الدراسات السابقة التي ركزت على الإعدادات المنفصلة والمسافة هامينغ. يقترح المؤلفون خوارزمية تعلم مناسبة وغير متحيزة للدوال المحدبة ليبشيتز، والتي تحقق تعقيد عينة قدره n^{O(1/ε^2)} ويؤكدون على ضرورة فرضية السلاسة لاستنتاج المحدبية من عينات محدودة.

El artículo aborda el aprendizaje y la prueba de funciones convexas de valores reales en un espacio gaussiano. Destaca las limitaciones de estudios previos que se centraron en configuraciones discretas y la distancia de Hamming. Los autores proponen un algoritmo de aprendizaje adecuado y agnóstico para funciones convexas Lipschitz, que logra una complejidad de muestra de n^{O(1/ε^2)} y enfatizan la necesidad de una suposición de suavidad para inferir la convexidad a partir de muestras finitas.

L'article traite de l'apprentissage et du test des fonctions convexes à valeurs réelles dans un espace gaussien. Il souligne les limites des études précédentes qui se concentraient sur des contextes discrets et la distance de Hamming. Les auteurs proposent un algorithme d'apprentissage approprié et agnostique pour les fonctions convexes Lipschitz, atteignant une complexité d'échantillonnage de n^{O(1/ε^2)} et soulignant la nécessité d'une hypothèse de régularité pour inférer la convexité à partir d'échantillons finis.

The article discusses the learning and testing of real-valued convex functions over Gaussian space. It highlights the limitations of previous studies that focused on discrete settings and the Hamming distance. The authors propose an agnostic proper learning algorithm for Lipschitz convex functions, which achieves a sample complexity of n^{O(1/ε^2)} and emphasizes the necessity of a smoothness assumption for inferring convexity from finite samples.

Event-Driven Digital-Time-Domain Inference Architectures for Tsetlin Machines

Was this article worth reading? Share it