arXiv:2509.12728v3 Announce Type: replace-cross 
Abstract: Phase retrieval in inline holography is a fundamental yet ill-posed inverse problem due to the nonlinear coupling between amplitude and phase in coherent imaging. We present a novel off-the-shelf solution that leverages a diffusion model trained solely on object amplitude to recover both amplitude and phase from diffraction intensities. Using a predictor-corrector sampling framework with separate likelihood gradients for amplitude and phase, our method enables complex field reconstruction without requiring ground-truth phase data for training. We validate the proposed approach through extensive simulations and experiments, demonstrating robust generalization across diverse object shapes, imaging system configurations, and modalities, including lensless setups. Notably, a diffusion prior trained on simple amplitude data (e.g., polystyrene beads) successfully reconstructs complex biological tissue structures, highlighting the method's adaptability. This framework provides a cost-effective, generalizable solution for nonlinear inverse problems in computational imaging, and establishes a foundation for broader coherent imaging applications beyond holography.

تم تقديم طريقة جديدة لاسترجاع الطور في الهولوجرافيا، باستخدام نموذج انتشار تم تدريبه فقط على سعة الجسم. تتيح هذه الطريقة استرجاع كل من السعة والطور من كثافات الانكسار دون الحاجة إلى بيانات الطور الحقيقية. تم التحقق من صحة الطريقة من خلال محاكاة وتجارب شاملة، مما يدل على فعاليتها عبر أشكال مختلفة من الأجسام وتكوينات التصوير، بما في ذلك الأنسجة البيولوجية. تقدم هذه الخطوة حلاً فعالاً من حيث التكلفة لمشاكل الانعكاس غير الخطية في التصوير الحسابي.

Se ha introducido un nuevo método para la recuperación de fase en holografía, utilizando un modelo de difusión entrenado únicamente en la amplitud del objeto. Este enfoque permite reconstruir tanto la amplitud como la fase a partir de intensidades de difracción sin necesidad de datos de fase de verdad. El método ha sido validado a través de simulaciones y experimentos extensos, demostrando su efectividad en diversas formas de objetos y configuraciones de imagen, incluyendo tejidos biológicos. Este avance ofrece una solución rentable para problemas inversos no lineales en la imagen computacional.

Une nouvelle méthode de récupération de phase en holographie a été introduite, utilisant un modèle de diffusion entraîné uniquement sur l'amplitude de l'objet. Cette approche permet de reconstruire à la fois l'amplitude et la phase à partir des intensités de diffraction sans nécessiter de données de phase de vérité terrain. La méthode a été validée par des simulations et des expériences approfondies, démontrant son efficacité à travers diverses formes d'objets et configurations d'imagerie, y compris les tissus biologiques. Cette avancée offre une solution rentable pour les problèmes inverses non linéaires en imagerie computationnelle.

A novel method for phase retrieval in holography has been introduced, utilizing a diffusion model trained solely on object amplitude. This approach allows for the reconstruction of both amplitude and phase from diffraction intensities without needing ground-truth phase data. The method has been validated through extensive simulations and experiments, demonstrating its effectiveness across various object shapes and imaging configurations, including biological tissues. This advancement offers a cost-effective solution for nonlinear inverse problems in computational imaging.

Generalizable Holographic Reconstruction via Amplitude-Only Diffusion Priors

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:2511.10941v1 Announce Type: new 
Abstract: Diffusion model (DM)-based channel estimation, which generates channel samples via a posteriori sampling stepwise with denoising process, has shown potential in high-precision channel state information (CSI) acquisition. However, slow sampling speed is an essential challenge for recent developed DM-based schemes. To alleviate this problem, we propose a novel flow matching (FM)-based generative model for multiple-input multiple-output (MIMO) channel estimation. We first formulate the channel estimation problem within FM framework, where the conditional probability path is constructed from the noisy channel distribution to the true channel distribution. In this case, the path evolves along the straight-line trajectory at a constant speed. Then, guided by this, we derive the velocity field that depends solely on the noise statistics to guide generative models training. Furthermore, during the sampling phase, we utilize the trained velocity field as prior information for channel estimation, which allows for quick and reliable noise channel enhancement via ordinary differential equation (ODE) Euler solver. Finally, numerical results demonstrate that the proposed FM-based channel estimation scheme can significantly reduce the sampling overhead compared to other popular DM-based schemes, such as the score matching (SM)-based scheme. Meanwhile, it achieves superior channel estimation accuracy under different channel conditions.

تقدم المقالة نموذجًا مبتكرًا قائمًا على مطابقة التدفق (FM) لتقدير قنوات MIMO. تتناول هذه الطريقة التحدي المتمثل في بطء سرعة أخذ العينات المرتبطة بالأنظمة القائمة على نماذج الانتشار (DM) من خلال صياغة مشكلة تقدير القناة ضمن إطار عمل FM. تُظهر الطريقة المقترحة إمكانية تحقيق دقة تقدير قنوات متفوقة وتقليل كبير في عبء أخذ العينات مقارنة بالطرق القائمة على DM الموجودة.

El artículo presenta un nuevo modelo generativo basado en el emparejamiento de flujos (FM) para la estimación de canales MIMO. Este enfoque aborda el desafío de la lenta velocidad de muestreo asociada con los esquemas basados en modelos de difusión (DM) al formular el problema de estimación de canales dentro del marco de FM. El método propuesto muestra un potencial para una precisión de estimación de canales superior y una reducción significativa de la sobrecarga de muestreo en comparación con los métodos DM existentes.

L'article présente un nouveau modèle génératif basé sur l'appariement de flux (FM) pour l'estimation de canaux MIMO. Cette approche répond au défi de la lenteur de la vitesse d'échantillonnage associée aux schémas basés sur les modèles de diffusion (DM) en formulant le problème d'estimation de canal dans le cadre du FM. La méthode proposée montre un potentiel pour une précision d'estimation de canal supérieure et une réduction significative de la surcharge d'échantillonnage par rapport aux méthodes DM existantes.

The article presents a novel flow matching (FM)-based generative model for multiple-input multiple-output (MIMO) channel estimation. This approach addresses the slow sampling speed challenge associated with diffusion model (DM)-based schemes by formulating the channel estimation problem within the FM framework. The proposed method shows potential for superior channel estimation accuracy and significantly reduced sampling overhead compared to existing DM-based methods.

Generalizable Holographic Reconstruction via Amplitude-Only Diffusion Priors

Was this article worth reading? Share it