arXiv:2604.11539v2 Announce Type: replace Abstract: Human perception of visual similarity is inherently adaptive and subjective, depending on the users' interests and focus. However, most image retrieval systems fail to reflect this flexibility, relying on a fixed, monolithic metric that cannot incorporate multiple conditions simultaneously. To address this, we propose CLAY, an adaptive similarity computation method that reframes the embedding space of pretrained Vision-Language Models (VLMs) as a text-conditional similarity space without additional training. This design separates the textual conditioning process and visual feature extraction, allowing highly efficient and multi-conditioned retrieval with fixed visual embeddings. We also construct a synthetic evaluation dataset CLAY-EVAL, for comprehensive assessment under diverse conditioned retrieval settings. Experiments on standard datasets and our proposed dataset show that CLAY achieves high retrieval accuracy and notable computational efficiency compared to previous works.
Science Journals
arXiv:2603.14711v2 Announce Type: replace Abstract: During the COVID-19 crisis, policymakers have implemented social bubble merging strategies, which allowed people from different households to meet and interact. Although these measures can mitigate the negative effects of extreme isolation, they also introduce additional contacts that may facilitate disease spread. As a result, several modeling studies have explored the epidemiological impact of different household-merging strategies, in which the selection of households to be merged is guided by specific demographic criteria, such as household size or the age composition of their members. Here, we investigate an alternative pairing strategy in which households are merged according to the number of economically active (working) members. We develop a mathematical model of household networks using real demographic data from multiple regions around the world, and simulate a lockdown scenario in which only economically active individuals can leave their households, while the remaining non-working members stay indoors. By using numerical simulations and the generating function technique, we then estimate the epidemic risk for different household merging strategies. We find that merging strategies based on the number of working members can keep epidemic risk at similar levels as those based on household size. Moreover, the worker-based approach allows significantly more people to form larger social bubbles, exceeding 40\% of the population in some countries. We find that merging households with at most one worker provides the best balance between controlling epidemic risk and addressing people's need for social contact.
arXiv:2604.19311v2 Announce Type: replace Abstract: We demonstrate a spatially resolved method for imaging millimeter-wave (mmWave) electric fields using Rydberg-state fluorescence in a warm atomic vapor. By utilizing a multi-photon ladder excitation scheme, we leverage a specific decay channel that remains dark in the absence of the mmWave field, resulting in high-contrast imaging with effectively zero background. Absolute calibration of the local electric field is achieved by reconstructing the Autler-Townes splitting of the Rydberg resonance across the imaging volume. To ensure robust field extraction across a wide dynamic range--including regimes where spectral features are not fully resolved--we employ a steady-state analysis based on the Gorini-Kossakowski-Sudarshan-Lindblad (GKSL) master equation. We apply this technique to visualize standing-wave interference patterns within a vapor cell and demonstrate the ability to engineer local field distributions using structured dielectric reflectors. This approach provides a versatile and self-calibrating platform for the diagnostic imaging of high-frequency electromagnetic fields and the characterization of mmWave-optical interfaces.
arXiv:2607.05766v1 Announce Type: new Abstract: A high resolution, Diagnostic Instrument Manipulator (DIM)-based x-ray Bragg crystal spectrometer has been calibrated for and deployed at the National Ignition Facility (NIF) to diagnose plasma conditions in ignition capsules near stagnation times. The spectrometer has two conical crystals in the Hall geometry focusing rays from the Kr He$\alpha$, Ly$\alpha$, and He$\beta$ complexes onto a streak camera, with the physics objectives of measuring time-resolved electron density and temperature through observing Stark broadening and the relative intensities of dielectronic satellites. A third von H\'amos crystal that time-integrates the Kr He$\alpha$, He$\beta$ and intervening energy range provides in-situ calibration for the streak camera signals. The spectrometer has been absolutely calibrated using a microfocus x-ray source, an array of CCD and single-photon-counting detectors, and multiple K- and L-absorption edge filters at the Princeton Plasma Physics Laboratory (PPPL) x-ray laboratory. Measurements of the integrated reflectivity, energy range, and energy resolution for each crystal are discussed. These calibration data provide absolute x-ray signal levels for NIF measurements, enabling precise filter selection and comparisons to simulations.
arXiv:2504.05294v3 Announce Type: replace Abstract: Chain-of-thought explanations are widely used to inspect the decision process of large language models (LLMs) and to evaluate the trustworthiness of model outputs, making them important for effective collaboration between LLMs and humans. We demonstrate that preference optimization - a key step in the alignment phase - can inadvertently reduce the faithfulness of these explanations. This occurs because the reward model (RM), which guides alignment, is tasked with optimizing both the expected quality of the response and the appropriateness of the explanations (e.g., minimizing bias or adhering to safety standards), creating potential conflicts. The RM lacks a mechanism to assess the consistency between the model's internal decision process and the generated explanation. Consequently, the LLM may engage in "reward hacking" by producing a final response that scores highly while giving an explanation tailored to maximize reward rather than accurately reflecting its reasoning. To address this issue, we propose enriching the RM's input with a causal attribution of the prediction, allowing the RM to detect discrepancies between the generated self-explanation and the model's decision process. In controlled settings, we show that this approach reduces the tendency of the LLM to generate misleading explanations.
arXiv:2607.04994v2 Announce Type: replace Abstract: The interaction of ultraintense lasers with magnetized overdense plasmas reveals a fundamentally new regime of laser-driven particle acceleration. Particle-in-cell simulations demonstrate the generation of directional, quasi-monoenergetic electrons in the MeV energy range superimposed on a broad thermal electron background with the estimated acceleration gradient of 3.6 MeV/{\mu}m, which is the highest till date. In contrast to conventional laser-plasma accelerators, which rely on underdense plasmas and are therefore constrained to relatively low plasma densities and limited beam charge, the present scheme operates in plasmas with densities orders of magnitude higher, opening new possibilities for the generation of high-flux energetic electron beams. A central result of this work is the demonstration of the excitation of electron Bernstein waves during relativistic laser interaction with magnetized overdense plasmas. The subsequent Landau damping of these electrostatic warm-plasma modes selectively transfers energy to resonant electrons, leading to the emergence of quasi-monoenergetic spectral peaks at energies that can be tuned through the applied magnetic field. To support the simulation results, we experimentally demonstrate the directional emission of energetic electrons from a simple permanent-magnet target irradiated by an ultraintense laser pulse, highlighting the practical feasibility of controlled electron-beam generation in dense plasma environments. These findings establish electron Bernstein waves as an efficient mediator of laser energy coupling in overdense plasmas and introduce a new paradigm for controlled particle acceleration and energy deposition in high-energy-density plasma systems.
arXiv:2604.06683v2 Announce Type: replace Abstract: Software architecture serves as the blueprint of a software system, capturing high-level structural decisions that shape downstream implementation and system quality. Despite this central role, generating architecture designs from requirement documents remains underexplored, with limited task-specific benchmarks for generation or rigorous evaluation. To bridge this gap, we introduce R2ABench, a benchmark for requirements-to-architecture (R2A) reasoning. R2ABench contains 68 projects with human-validated references collected from both educational-style settings and GitHub repositories. Each project is packaged as a unified instance that provides a structured software requirements specification (SRS), a reference architecture view in PlantUML, and the corresponding source requirement artifacts. To evaluate generated architecture views, we design a layered hybrid evaluation framework spanning syntax validation, structural graph diagnostics, and semantic and evidence-based architecture scoring. Using this framework, we conduct a comprehensive empirical study of state-of-the-art LLMs and agents on R2ABench. We find that current systems can often generate syntactically valid and readable architecture views, but still struggle with relation-level architecture modeling across both subsets: component identification is substantially stronger than edge recovery, and edge hallucination is the dominant structural failure mode. Semantically, structural fidelity does not guarantee requirement coverage or traceability, which emerge as the primary quality gaps. The associated data files are available at https://figshare.com/s/01f0a5fb6243a6a60f23.
arXiv:2607.06133v1 Announce Type: new Abstract: Modern software systems increasingly depend on data for analysis, prediction, testing, and decision-making. Yet many important domains, including medicine, safety-critical systems, and regulated industries, lack abundant, shareable, or representative data. Synthetic data generation is often proposed as a remedy, but our experience engineering software for intraoperative radiotherapy (IORT) in breast cancer treatment suggests that synthetic data shifts rather than solves the central engineering problem. The key challenge becomes deciding which properties synthetic data must preserve, how these properties should be elicited from stakeholders, how they can be validated under privacy constraints, and how they evolve. We call this problem property-driven synthetic data engineering. Drawing on a collaboration with oncologists and preliminary experiments with a sensitive IORT dataset, we identify challenges in requirements, validation, privacy, and pipeline evolution. We argue that automated software engineering research should develop methods and tools for eliciting, formalizing, checking, and evolving validity properties for synthetic data in data-scarce software systems.
arXiv:2607.06145v1 Announce Type: new Abstract: In this paper, we define the quantity of prompting complexity: for a fixed instruction-tuned language model, what is the shortest plausible prompt that makes deterministic decoding produce a target text? It is an LM-relative analogue of resource-bounded Kolmogorov complexity: the prompt is a program, the model interface is the interpreter, and information omitted from the prompt is supplied by the model's weights, training distribution, tokenizer, template, and decoding rule. Unlike classical Kolmogorov complexity, this measure is intentionally non-universal. In the finite-context setting it is computable by enumeration, but there is no model-independent invariance theorem; the same text may be cheap for one model and inaccessible or expensive for another. To keep the search space aligned with prompt engineering, we restrict programs to plausible human-readable texts rather than arbitrary token strings. We extend the exact definition to soft prompting complexity for approximate outputs, yielding a lossy notion of model-relative text compression and a formal target for prompt optimization. We also define prompting distance by comparing shortest generating prompts, and behavioral prompting complexity for reaching any output satisfying a specification. Based on these formulations, we define a research agenda for empirically studying which texts and behaviors are accessible from short plausible prompts under a fixed LM interface.
arXiv:2604.04738v2 Announce Type: replace Abstract: Fine-tuning is the dominant paradigm for adapting large machine learning models, yet current deployment pipelines provide no way to verify how a released model was updated. In particular, a model provider or auditor cannot check whether a fine-tuned model adheres to a claimed update procedure without access to its parameters. We introduce \emph{fine-tuning integrity} (FTI), a cryptographic objective for verifying that a deployed model differs from a trusted base model only within a declared class of admissible updates. We construct \emph{succinct model difference proofs} (SMDPs), zero-knowledge protocols that certify structured parameter drift without revealing model weights. Our framework supports three fundamental update classes: norm-bounded, low-rank, and sparse drift, covering common fine-tuning methods such as regularized training, LoRA, and prefix tuning. In all cases, proof size and verification cost depend on the structure of the update rather than the number of parameters. We prove soundness, zero-knowledge, and succinctness for each construction, and establish a matching $\Omega(n)$ lower bound showing that structural assumptions are necessary for succinct verification. A prototype evaluation on synthetic benchmarks and GPT-2 fine-tuning demonstrates that proofs remain compact and verification is efficient at realistic scales.
arXiv:2604.21927v3 Announce Type: replace Abstract: Continual learning (CL) studies how models acquire tasks sequentially while retaining previously learned knowledge. Despite substantial progress in benchmarking CL methods, comparative evaluations typically keep the fine-tuning regime fixed. In this paper, we argue that the fine-tuning regime, defined by the trainable parameter subspace, is itself a key evaluation variable. We formalize adaptation regimes as projected optimization over fixed trainable subspaces, showing that changing the trainable depth alters the effective update signal through which both current task fitting and knowledge preservation operate. This analysis motivates the hypothesis that method comparisons need not be invariant across regimes. We test this hypothesis in task incremental CL, five trainable depth regimes, and four standard methods: online EWC, LwF, SI, and GEM. Across five benchmark datasets, namely MNIST, Fashion MNIST, KMNIST, QMNIST, and CIFAR-100, and across 11 task orders per dataset, we find that the relative ranking of methods is not consistently preserved across regimes. We further show that deeper adaptation regimes are associated with larger update magnitudes, higher forgetting, and a stronger relationship between the two. These results show that comparative conclusions in CL can depend strongly on the chosen fine-tuning regime, motivating regime-aware evaluation protocols that treat trainable depth as an explicit experimental factor.
arXiv:2607.06018v1 Announce Type: new Abstract: Pretrained video generative models are promising backbones for visuomotor control, but their imagined futures often drift from task intent and are not reliably action-conditional. As a result, these models can be difficult to use for planning or policy extraction. To address these limitations, we propose RoboTALES, a single-stage framework that learns task-aligned simulated futures and uses them to train robot policies. Our approach introduces two key innovations: (1) a hierarchical LLM-based planner that breaks complex tasks into a sequence of subgoals to guide the model's imagination; and (2) a VLM-based critic that evaluates these ``imagined'' futures and uses reward-based feedback to keep the model's internal representations focused on the goal. By anchoring the video generator in abstract reasoning, we produce temporally consistent rollouts and more coherent actions. We evaluate RoboTALES on diverse manipulation tasks from RoboCasa and LIBERO10, and show that our method consistently outperforms existing methods, especially in long-horizon tasks. Our code and models are publicly available at https://github.com/hananshafi/RoboTALES.
arXiv:2405.14678v2 Announce Type: replace-cross Abstract: In this article, we combine Sweedler's classic theory of measuring coalgebras -- by which $k$-algebras are enriched in $k$-coalgebras for $k$ a field -- with the theory of W-types -- by which the categorical semantics of inductive data types in functional programming languages are understood. In our main theorem, we find that under some hypotheses, algebras of an endofunctor are enriched in coalgebras of the same endofunctor, and we find polynomial endofunctors provide many interesting examples of this phenomenon. We then generalize the notion of initial algebra of an endofunctor using this enrichment, thus generalizing the notion of W-type. This article is an extended version of arXiv:2303.16793, it adds expository introductions to the original theories of measuring coalgebras and W-types along with some improvements to the main theory and many explicitly worked examples.
arXiv:2607.02383v2 Announce Type: replace Abstract: LLM-based retrieval-augmented generation (RAG) is increasingly used for automated fact-checking (AFC) and related tasks. By grounding LLM outputs in retrieved evidence, RAG-based systems provide transparent justifications while allowing external information to be updated independently of the underlying model. However, existing approaches often assume retrieved evidence is reliable, although real-world information may be conflicting, outdated, and can originate from unreliable or biased sources. Recent work on *source-critical reasoning* addresses this challenge through media background checks (MBCs) (Schlichtkrull, 2024), which assess the credibility of evidence sources to support downstream fact verification. However, generating MBCs relies on costly proprietary search APIs, limiting reproducibility. To mitigate this issue, we introduce MEDIAREF, a publicly available knowledge store of web-sourced documents that enables reproducible, low-cost evaluation of MBC generation across 200 media sources. We describe a reproducible methodology for constructing and updating the collection, assess widely used LLMs on the MBC generation task, and demonstrate that MEDIAREF supports higher-quality MBC generation through both automatic and qualitative evaluation.
arXiv:2512.23236v4 Announce Type: replace Abstract: Making deep learning recommendation model (DLRM) training and inference fast and efficient is important. However, this presents three key system challenges - model architecture diversity, kernel primitive diversity, and hardware generation and architecture heterogeneity. This paper presents KernelEvolve-an agentic kernel coding framework-to tackle heterogeneity at-scale for DLRM. KernelEvolve is designed to take kernel specifications as input and automate the process of kernel generation and optimization for recommendation model across heterogeneous hardware architectures. KernelEvolve does so by operating at multiple programming abstractions, from Triton and CuTe DSL to low-level hardware agnostic languages, spanning the full hardware-software optimization stack. The kernel optimization process is described as graph-based search with selection policy, universal operator, fitness function, and termination rule, dynamically adapts to runtime execution context through retrieval-augmented prompt synthesis. We designed, implemented, and deployed KernelEvolve to optimize a wide variety of production recommendation models across generations of NVIDIA and AMD GPUs, as well as Meta's AI accelerators. We validate KernelEvolve on the publicly-available KernelBench suite, achieving 100% pass rate on all 250 problems across three difficulty levels, and 160 PyTorch ATen operators across three heterogeneous hardware platforms, demonstrating 100% correctness. KernelEvolve reduces development time from weeks to hours and achieves substantial performance improvements over PyTorch baselines across diverse production use cases and for heterogeneous AI systems at-scale. Beyond performance efficiency improvements, KernelEvolve significantly mitigates the programmability barrier for new AI hardware by enabling automated kernel generation for in-house developed AI hardware.
arXiv:2601.05337v5 Announce Type: replace Abstract: We discuss, at a level accessible to students and teachers of physics and mathematics, the fundamental transition from classical mechanics to wave equations made by Schrodinger a century ago. These historical developments clarify the structure and significance of the quantum mechanics.
arXiv:2604.02537v2 Announce Type: replace Abstract: All-atom molecular dynamics (MD) simulations can predict polymer properties from molecular structure, yet their execution requires specialized expertise in force field selection, system construction, equilibration, and property extraction. We present PolyJarvis, an agent that couples a large language model (LLM) with established simulation toolkits, including Enhanced Monte Carlo (EMC) for system construction and LAMMPS for molecular dynamics, through Model Context Protocol (MCP) servers, enabling end-to-end polymer property prediction from natural language input. Given a polymer name or SMILES string, PolyJarvis orchestrates molecular model construction, equilibration, and thermal/mechanical property calculation. Validation is conducted on nine amorphous homopolymers spanning seven chemistries: polyethylene (PE), polystyrene (PS), poly(methyl methacrylate) (PMMA), poly(ethylene glycol) (PEG), poly(ether ether ketone) (PEEK), poly(vinyl chloride) (PVC), poly(lactic acid) (PLA), polysulfone (PSU), and cis-polybutadiene (cis-PBD). On the replicate mean over four runs, 18 of the 25 property comparisons with experimental references meet the acceptance criteria (glass transition within 50K, density within 5%, bulk modulus within 30%): glass transition 7 of 9, density 5 of 9, and bulk modulus 6 of 7. The failures fall into two groups: polymer consistent force field (PCFF) systems that run under-dense, and the rigid backbones PLA and PEEK, which overestimate the glass transition on cooling. Each was traced to a protocol or an analysis step of the workflow. As a proof of concept, this work shows that an LLM-driven agent can carry out end-to-end polymer MD workflows, with predictive accuracy that varies across properties and polymers.
arXiv:2607.06124v1 Announce Type: new Abstract: The increasing energy demand of software systems is raising concerns about their environmental impact and associated costs. Reasoning on energy usage early in the development flow has the potential to significantly reduce the overall energy usage of a software system, as it allows developers to make informed design and refactoring decisions before inefficiencies propagate. However, assessing energy usage without repeated profiling and direct measurement is difficult, which limits early reasoning in practice. This study investigates the limits of method-level energy prediction in Java, examining whether static source code metrics complemented with method-level execution time can estimate the energy consumption of Java methods. We profile 2,786 Java methods to extract 33 static features and measure execution time and energy, then train and compare eleven regression models. Our findings show that static source code metrics alone yield poor predictive performance, with average R2 values close to zero. Incorporating execution time as a lightweight dynamic input significantly improves accuracy, raising R2 to as high as 0.46. Execution time, internal method calls, and cyclomatic complexity consistently emerge as the strongest predictors of energy consumption.
arXiv:2607.06136v1 Announce Type: new Abstract: Recent diffusion-based generative models have shown impressive performance in image generation and editing. However, due to memory limitations and the high cost of collecting high-resolution training images, existing methods are typically restricted to inputs with linear resolutions below 1K. In contrast, photos captured by modern mobile devices often reach linear resolutions up to 8K, revealing a significant gap between current capabilities and real-world demands. Simply upscaling low-resolution edited results often results in visually enlarged but blurry images that lack fine details. This paper introduces UltraDiffEdit, a novel, tuning-free image editing framework that extends off-the-shelf latent diffusion models (LDMs) to ultrahigh resolutions. UltraDiffEdit employs a multi-scale progressive editing strategy, iteratively blending high-resolution edited content with unedited areas in a coarse-to-fine manner. We employ multi-patch encoding to preserve both edited and unedited visual details within the latent space. To mitigate editing artifacts, our global-local consistency denoising technique consistently integrates edited and unedited latent features, ensuring smooth transition at editing boundaries from the latent representation to the final image. We also introduce a patch-based hybrid sampling approach that captures local, intermediate, and global features, ensuring semantic coherence and enhancing fine detail during denoising. We conduct extensive experiments demonstrating UltraDiffEdit's superior editing quality and flexibility: it can handle image resolutions up to 8K using only a single NVIDIA GeForce RTX 3090 GPU. The source code is publicly available at https://github.com/LonglongaaaGo/UltraDiffEdit.
arXiv:2505.10356v3 Announce Type: replace Abstract: Decoding language from the human brain remains a grand challenge for Brain-Computer Interfaces (BCIs). Current approaches typically rely on unimodal brain representations, neglecting the brain's inherently multimodal processing. Inspired by the brain's associative mechanisms, where viewing an image can evoke related sounds and linguistic representations, we propose a unified framework that leverages Multimodal Large Language Models (MLLMs) to align brain signals with a shared semantic space encompassing text, images, and audio. A router module dynamically selects and fuses modality-specific brain features according to the characteristics of each stimulus. Experiments on various fMRI datasets with textual, visual, and auditory stimuli demonstrate state-of-the-art performance, achieving an 8.48% improvement on the most commonly used benchmark. We further extend our framework to EEG and MEG data, demonstrating flexibility and robustness across varying temporal and spatial resolutions. To our knowledge, this is the first unified BCI architecture capable of robustly decoding multimodal brain activity across diverse brain signals and stimulus types, offering a flexible solution for real-world applications.
arXiv:2607.04244v2 Announce Type: replace Abstract: This report describes our approach to the Efficient Qwen Competition, where the goal is to enable low-latency serving of Qwen3.5-4B on a resource-constrained NVIDIA A10G GPU. Our system combines a quantized target model with speculative decoding. To recover accuracy, we apply quantization-aware distillation to the target model while retaining the original quantization grid. To speed up decoding, a block-diffusion drafter specialized for the quantized target model is trained using a two-stage procedure: first learning from the high-precision target and then adapting to the low-precision target. Because the drafter is invoked at every speculative decoding step, we further reduce its overhead with quantization and sliding-window attention, preserving draft-token acceptance while improving long-context decoding latency. As a result, our submission achieves a 6.978$\times$ average speedup over the baseline while satisfying the required quality thresholds, ranking 3rd overall. We hope these results provide useful insights for practical LLM inference. The code and resources are available at https://github.com/nota-github/adaptfm-quant-dflash
arXiv:2607.06176v1 Announce Type: new Abstract: Scene graph generation (SGG) approaches can be broadly classified into detector-based and query-based methods according to their underlying reasoning mechanisms. However, the discrepancy in their predictive behaviors, induced by these distinct mechanisms, has not been systematically analyzed. In this work, we design a controlled experimental setup to examine prediction discrepancies from the perspective of detector-conditioned reachability. The results suggest clear complementary clues. Motivated by this observation, we introduce a Dual-SGG method that consolidates both reasoning mechanisms via a dual-query design, thereby leveraging the complementary predictive behaviors of both detector-based and query-based methods. Extensive experiments on the Visual Genome, Open Images v6, and GQA-200 datasets demonstrate the effectiveness of the proposed method.
arXiv:2607.06187v1 Announce Type: new Abstract: We introduce generalized altitudes of a simplex, extending the usual vertex-to-opposite-face altitude to arbitrary pairs of opposite faces. These quantities encode the relative position of the affine spans of such faces and yield a uniform formula for the angle between them. We also derive an equivalent algebraic expression in terms of generalized cross products and Gram determinants, linking the construction to standard determinant-based tools. Finally, we prove that every generalized altitude is bounded below by a quantity controlled by the ordinary height of the simplex. Thus, classical height or thickness assumptions imply control over this broader family of geometric quantities. The results provide a compact framework for studying simplex quality and are motivated by applications to triangulation criteria for Riemannian manifolds.
arXiv:2603.19216v2 Announce Type: replace Abstract: Understanding and generating 3D objects as compositions of meaningful parts is fundamental to human perception and reasoning. However, most text-to-3D methods overlook the semantic and functional structure of parts. While recent part-aware approaches introduce decomposition, they remain largely geometry-focused, lacking semantic grounding and failing to model how parts align with textual descriptions or their inter-part relations. We propose DreamPartGen, a framework for semantically grounded, part-aware text-to-3D generation. DreamPartGen introduces Duplex Part Latents (DPLs) that jointly model each part's geometry and appearance, and Relational Semantic Latents (RSLs) that capture inter-part dependencies derived from language. A synchronized co-denoising process enforces mutual geometric and semantic consistency, enabling coherent, interpretable, and text-aligned 3D synthesis. Across multiple benchmarks, DreamPartGen delivers state-of-the-art performance in geometric fidelity and text-shape alignment.
arXiv:2607.03261v2 Announce Type: replace Abstract: Recent large language models (LLMs) have demonstrated remarkable progress in 3D spatial reasoning, spatial grounding, and fine-grained geometric understanding. However, their ability to reason about densely packed object placement under strict spatial and functional constraints remains largely unexplored, despite being a fundamental challenge in practical electronic design automation (EDA) workflows. To bridge this gap, we introduce OmniLayout, the first benchmark designed to evaluate LLMs on printed-circuit-board (PCB) layout placement reasoning under real-world geometric, routing, and connectivity constraints. OmniLayout contains 1,681 industrial-grade schematic-coupled PCB layouts and includes four tasks: (1) geometric reasoning for IC physical placement, with 77.24K placement instances constrained within PCB board boundaries; (2) routability-aware placement reasoning, generating physically valid component placements; (3) electrical functionality, preserving schematic-specified connectivity and electronic functional correctness; and (4) tool-augmented agentic reasoning for invoking external tools to accomplish tasks (1)-(3). Our results reveal substantial limitations of current LLMs in PCB layout placement, including weak geometric reasoning, poor routability optimization, and inconsistent preservation of electrical functionality.