arXiv:2606.05576v1 Announce Type: new
Abstract: Vision-language models (VLMs) excel on visual question answering and multimodal reasoning benchmarks. Yet their capability on ultra-resolution images - where critical evidence is tiny, subtle, spatially distant, or distributed - remains unclear. Existing evaluations largely report final-answer accuracy, offering limited insight into whether models acquire and integrate the necessary visual evidence. We introduce UltraVR, a diagnostic benchmark for evidence-grounded visual reasoning over ultra-resolution images. UltraVR spans four high-value scenarios: CCTV surveillance, remote sensing (RS), whole-slide image (WSI) pathology, and industrial anomaly detection (AD). These domains pose complementary challenges: fine-grained object grounding in crowded CCTV scenes, long-range spatial comparison in RS, multi-scale evidence navigation in WSI, and subtle irregularity detection in repetitive industrial layouts. Beyond standard QA triples, each instance includes a structured ground-truth chain of thought with step-level questions, intermediate answers, and reasoning labels. These labels decompose reasoning into evidence grounding, local perception, quantification, evidence integration, and decision inference, enabling process-level diagnosis over black-box scoring. Using UltraVR, we evaluate frontier VLMs and show that current models remain far from reliable on ultra-resolution reasoning. Importantly, the structured annotations allow us to localize failures across the visual-to-decision pipeline: errors concentrate in evidence grounding and local perception, while downstream inference often recovers when intermediate visual facts are supplied. These findings demonstrate UltraVR as a diagnostic testbed for measuring not only whether VLMs answer correctly, but where their ultra-resolution reasoning process breaks.
Science Journals
arXiv:2606.06320v1 Announce Type: new
Abstract: Machine unlearning aims to remove targeted knowledge from a trained model while preserving its general capabilities. For autoregressive language models, not all tokens in a forget sample are equally relevant to forgetting. Existing approaches either ignore this heterogeneity or rely on auxiliary models, heuristics, or external annotations to estimate each token's relevance for forgetting. We instead characterize it through the interaction with the retain objective: a token is forget-specific to the extent that minimizing the forget loss on that token does not conflict with retain optimality. We formalize this perspective as a joint optimization problem over the model parameters and the token weights and show that, under a natural separation condition, the resulting objective recovers the oracle forget-specific token support. Motivated by this formulation, we introduce Alternating Token-Weighted Unlearning (ATWU), a lightweight framework that jointly learns token forget-specificity and model parameters during unlearning using a simple linear scorer over the hidden states, without external token level supervision. Across TOFU and RWKU, ATWU achieves state of the art forget-retain trade-offs, outperforming sample-level methods, probability-based token weighting heuristics, and auxiliary-model-based approaches. Moreover, the learned scores align substantially better with ground truth forget-specific spans, indicating that ATWU identifies semantically meaningful token level forgetting signals. Overall, our results suggest that retain conflict provides an effective criterion for identifying what language models should forget, enabling unsupervised learning of token level forget-specificity directly from model representations with minimal computational overhead.
arXiv:2606.06314v1 Announce Type: new
Abstract: Time-dependent high-dimensional partial differential equations (PDEs) with spatially localised and dynamically evolving solutions pose a fundamental challenge for physics-informed neural networks (PINNs), as uniform collocation sampling becomes increasingly ineffective in high-dimensional spatiotemporal domains. In this work, a deep adaptive sampling framework for PINNs is extended to the time-dependent setting by treating space and time as a unified domain without any explicit time marching. A normalising flow neural network model effectively learns the distribution induced by the PDE residual and generates new collocation points concentrated in regions where the solution is most difficult to learn. Unlike conventional adaptive strategies that require explicit time stepping or moving meshes, high-residual regions are automatically identified and tracked across both space and time, driven purely by the PDE residual distribution. The effectiveness of the proposed strategy is assessed on a range of benchmark problems, from sharp and moving features in two spatial dimensions to localised structures in up to eight spatial dimensions.
arXiv:2606.06300v1 Announce Type: new
Abstract: We propose MResOpt, a staged residual neural network architecture for constrained optimization problems. Our architecture fits within predict-complete-correct pipelines and decomposes constraint satisfaction by priority through intermediate re-completion and stage-aware losses. The framework enables domain-informed ordered constraint satisfaction which allows the network to utilize ordinal structure when present. Under an idealized infinite-width regime, we show that our design behaves as sequential Gaussian Process regression. On synthetic QP, QCQP, and SOCP benchmarks, the staged architecture improves high-priority constraint satisfaction across convex and non-convex settings. On line-flow-constrained AC optimal power flow, we introduce a physics-motivated constraint ordering and show that MResOpt supports a learned division of labor that keeps iterates on the equality manifold, achieving substantially lower high-priority violation than reprojected baselines while remaining computationally efficient.
arXiv:2606.06297v1 Announce Type: new
Abstract: In the fields of vision science, cognitive psychology, and psycholinguistics, the accurate measurement of reading speed is frequently hampered by the limitations of static reading charts. Repeated testing often leads to memorization effects, while the requirement for oral recitation introduces speech-motor confounds that obscure true information processing speed. To address these methodological hurdles, this paper introduces an open-source MATLAB toolbox that adapts the sentence generation paradigm originally proposed by Perrin, Paill\'e, and Baccino (2014) for the English language. This system utilizes a semantic ontology and a "proto-truth" logic to autonomously generate thousands of unique, grammatically simple sentences with unambiguous truth values. Beyond the original scope of Maximum Reading Speed (MRS) measurement, this implementation introduces band-pass psycholinguistic filtering and specific logic to resolve semantic ambiguities unique to English. We present this complete software package as an open platform for the scientific community to validate and refine.
arXiv:2606.06292v1 Announce Type: new
Abstract: Robotic manipulation of textiles remains challenging because continuous deformation and self-occlusions hinder the robust visual perception required to estimate the cloth's state. To address the lack of annotated real-world data, we developed a Blender-based synthetic pipeline exporting auto-annotated keypoints, and combined manually labeled renders with real-world data to train a wrinkle detector. We present a perception framework integrating a CNN for permutation-invariant keypoint detection and a YOLOv8-OpenCV pipeline to extract grasping points from structural wrinkles. A proposed bimanual algorithm uses this system to stretch fully folded garments via wrinkles, transitioning to keypoint-based ironing once corners emerge. The keypoint model achieves a Mean Position Error (MPE) of 1.7615 pixels. The perception system transfers to physical fabrics without fine-tuning, outperforming baselines that fail in high-occlusion states or yield false positives on severe folds.
arXiv:2606.06286v1 Announce Type: new
Abstract: Large language models can reproduce training data, but existing memorization evaluations mostly measure whether models can be forced to do so, rather than whether they do so under ordinary use. We introduce PropMe, a propensity-aware framework for memorization evaluation that contrasts prefix-based capability attacks with non-adversarial evaluations. We propose a metric transformation that, applied to existing functions, allows to create propensity metrics. We further introduce SimpleTrace, a lightweight tracing pipeline built on infini-gram that deterministically attributes model generations to large-scale training corpora and computes verbatim, near-verbatim, and propensity-transformed memorization metrics. Evaluating two fully-open models: Comma and DFM Decoder on two datasets: Common Pile and Dynaword in two languages, we find a consistent gap between capability and propensity: prefix attacks elicit substantially stronger memorization signals than generic or dataset-specific prompts, while propensity scores remain low overall. Thus, the models can reveal training data when directly elicited, but rarely do so in more common non-adversarial settings. We also find that DFM Decoder, which is continually pre-trained from Comma, exhibits reduced memorization and memorization propensity for Common Pile, confirming that memorization capability can decrease when later training emphasizes partially different data. Our results suggest, and we encourage, that memorization audits should report both worst-case extractability and ordinary leakage propensity in order to have a more comprehensive view of this phenomenon.
arXiv:2602.16705v3 Announce Type: replace
Abstract: Visual loco-manipulation of arbitrary in-the-wild objects requires accurate end-effector (EE) control and a generalizable understanding of the scene from visual inputs (eg, RGB-D images). Existing imitation and sim2real methods jointly learn both these aspects via monolithic end-to-end learning and are thus hard to scale. In this work, we bring to bear the best tools for each of these problems -- large vision models for generalizable scene understanding and simulated training for accurate EE control -- leading to an overall modular loco-manipulation system that exhibits strong generalization. Our core technical innovation is HERO, an accurate residual-aware EE tracking policy made possible by combining classical robotics with machine learning. It uses a) inverse kinematics to convert residual end-effector targets into reference trajectories, b) a learned neural forward model for accurate forward kinematics, and c) goal adjustment and replanning. Together, these innovations reduce the end-effector tracking error to 2.44cm, outperforming the strongest prior method by 5.5x. Our overall system operates in diverse real-world environments, from offices to coffee shops, where the robot reliably grasps various everyday objects (eg, mugs, apples, toys) on surfaces ranging from 43cm to 92cm in height. Systematic modular and end-to-end tests demonstrate the effectiveness of our proposed design. We believe our advances open up new ways of training humanoids to interact with daily objects.
arXiv:2606.06278v1 Announce Type: new
Abstract: Blind image restoration requires recovering clean images from observations corrupted by unknown and potentially mixed degradations. While recent deterministic flow-based methods model restoration as transport processes that map degraded images to clean ones, they typically rely on Euclidean interpolation, implicitly assuming linear degradation geometry. In this paper, we explicitly model degradations as points on a low-dimensional Riemannian manifold and formulate restoration as geodesic transport on the joint image-manifold space. Using a geodesic flow matching objective, we learn intrinsic transport dynamics that respect the curvature of degradation space. This framework generalizes linear flow matching, provides a principled treatment of mixed degradations as geodesic compositions, and yields a clean theoretical interpretation for generalization beyond observed degradations.
arXiv:2606.06275v1 Announce Type: new
Abstract: Student engagement is critical for effective learning in software modelling, yet fostering motivation and inclusivity remains a challenge. While existing research has focused on modelling tools, notations, and assessment, little attention has been given to how the choice of problem domains and the diversity, relatability, and cultural perspectives they bring shape students' learning experiences. This study explores how problem domains and teaching methods influence motivation, engagement, inclusiveness, and feedback in modelling education. To investigate these dimensions, we conducted parallel surveys with 90 students and 22 educators. Our findings reveal disconnects between educator assumptions and student preferences: Students show greatest motivation for socially relevant domains and prefer choice in selection, while educators overestimate interest in study-related domains. The study identifies how minor design choices can exclude students. Students perceive feedback as meaningful when visibly acted upon. These findings suggest inclusive domain selection is central to student motivation; thus, we recommend student-centred domain selection.
arXiv:2603.14210v2 Announce Type: replace
Abstract: We present Vavanagi, a community-run platform for Hula (Vula'a), an Austronesian language of Papua New Guinea with approximately 10,000 speakers. Vavanagi supports crowdsourced English-Hula text translation and voice recording, with elder-led review and community-governed data infrastructure. To date, 77 translators and 4 reviewers have produced over 12k parallel sentence pairs covering 9k unique Hula words. We also propose a multi-level framework for measuring community involvement, from consultation to fully community-initiated and governed projects. We position Vavanagi at Level 5: initiative, design, implementation, and data governance all sit within the Hula community, making it, to our knowledge, the first community-led language technology initiative for a language of this size. Vavanagi shows how language technology can bridge village-based and urban members, connect generations, and support cultural heritage on the community's own terms.
arXiv:2606.06272v1 Announce Type: new
Abstract: Generative Flow Networks (GFlowNets) are a framework for sampling structured objects via stochastic trajectories in a directed graph. In this work, we establish a theoretical connection between non-acyclic GFlowNets and optimal transport (OT). We show that fixing the initial flow distribution in a minimum-flow GFlowNet reduces its objective to a Kantorovich OT problem with graph-induced shortest path costs. At the optimum, the learned GFlowNet policy therefore encodes an optimal transport plan from the source distribution to the target distribution: we show that sampling trajectories from the minimum-flow GFlowNet recovers the corresponding optimal coupling. Our formulation enables applying the GFlowNet learning framework to OT problems on large graphs via edge flows and neural parameterization. Experiments confirm agreement with exact OT solvers and demonstrate that GFlowNets can learn high-quality transport plans.
arXiv:2603.14805v2 Announce Type: replace
Abstract: Enterprise software organizations accumulate critical institutional knowledge - architectural decisions, deployment procedures, compliance policies, incident playbooks - yet this knowledge remains trapped in formats designed for human interpretation. The bottleneck to effective agentic software development is not model capability but knowledge architecture. When any knowledge consumer - an autonomous AI agent, a newly onboarded engineer, or a senior developer - encounters an enterprise task without institutional context, the result is guesswork, correction cascades, and a disproportionate tax on senior engineers who must manually supply what others cannot infer.
This paper introduces Knowledge Activation, a framework that specializes AI Skills - the open standard for agent-consumable knowledge - into structured, governance-aware Atomic Knowledge Units (AKUs) for institutional knowledge delivery. Rather than retrieving documents for interpretation, AKUs deliver action - ready specifications encoding what to do, which tools to use, what constraints to respect, and where to go next - so that agents act correctly and engineers receive institutionally grounded guidance without reconstructing organizational context from scratch.
AKUs form a composable knowledge graph that agents traverse at runtime - compressing onboarding, reducing cross - team friction, and eliminating correction cascades. The paper formalizes the resource constraints that make this architecture necessary, specifies the AKU schema and deployment architecture, and grounds long - term maintenance in knowledge commons practice.
A Yahoo deployment surveying 67 engineers shows statistically significant developer-experience gains - 2.6 hours per week saved, Net Promoter Score +35. Organizations that architect their institutional knowledge for the agentic era will outperform those that invest solely in model capability.
arXiv:2606.06267v1 Announce Type: new
Abstract: Circuit discovery methods identify subgraphs that explain specific model behaviors, and structural differences between discovered circuits are commonly interpreted as evidence of distinct mechanisms. We test this assumption by varying input statistics while holding the task fixed, and show that the resulting structural differences exhibit apparent specialization but do not correspond to functional differences, a pattern we term phantom specialization. Using Literal Sequence Copying across four token-frequency bands plus a control condition in five Pythia models (70M-1.4B), we extract 75 circuits and find that structurally distinct circuits implement the same computation: band-specific edges transfer broadly across bands, a core shared across most bands recovers at least 99% of circuit performance, and causal interchange interventions confirm that internal representations are interchangeable across frequency bands. Repeated extractions within the same frequency band further suggest that discovery algorithms sample from an equivalence class of valid subgraphs rather than recovering a unique mechanism. Standard evaluation practice obscures this pattern: source-level evaluation inflates apparent faithfulness, while edge-level evaluation reveals the many-to-one mapping from structure to function. Our results show that structural differences between circuits are not sufficient evidence for distinct mechanisms, and that exposing this requires edge-level evaluation and cross-condition transfer tests.
arXiv:2603.16475v2 Announce Type: replace
Abstract: In schema-guided reasoning (SGR) pipelines, LLMs produce explicit intermediate structures -- rubrics, checklists, or verification queries -- before committing to a final decision. SGR is increasingly adopted because it promises controllability: practitioners expect to inspect, edit, and override these structures to steer the outcome. But does the promise hold? We introduce a causal evaluation protocol to measure it: by selecting tasks where a deterministic function maps intermediate structures to decisions, every controlled edit implies a unique correct output. Across 12 models and 4 benchmarks, models appear self-consistent with their own intermediate structures but fail to update predictions after intervention -- revealing that apparent faithfulness is fragile once the intermediate structure changes. When derivation of the final decision from the structure is delegated to an external tool, this fragility largely disappears; stronger prompting yields only limited improvements, while preference optimization substantially improves intervention faithfulness. Overall, intermediate structures in schema-guided pipelines function as influential context rather than stable causal mediators.
arXiv:2604.03042v2 Announce Type: replace
Abstract: Multi-agent autonomous exploration is essential for applications such as environmental monitoring, search and rescue, and industrial-scale surveillance. However, effective coordination under communication constraints remains a significant challenge. Frontier exploration algorithms analyze the boundary between the known and unknown regions to determine the next-best view that maximizes exploratory gain. This article proposes an enhancement to existing frontier-based exploration algorithms by introducing a probabilistic approach to frontier prioritization. By leveraging Dirichlet process Gaussian mixture model (DP-GMM) and a probabilistic formulation of information gain, the method improves the quality of frontier prioritization. The proposed enhancement, integrated into two state-of-the-art multi-agent exploration algorithms, consistently improves performance across environments of varying clutter, communication constraints, and team sizes. Simulations showcase an average gain of $10\%$ and $14\%$ for the two algorithms across all combinations. Successful deployment in real-world experiments with a dual-drone system further corroborates these findings.
arXiv:2604.03634v5 Announce Type: replace
Abstract: We establish that temporal averaging over multiple observations is the degenerate case of algebraic group action with the trivial group $G=\{e\}$. A General Replacement Theorem proves that a group-averaged estimator from one snapshot achieves equivalent subspace decomposition to multi-snapshot covariance estimation. The Trivial Group Embedding Theorem proves that the sample covariance is the accumulation of trivial-group estimates, with variance governed by a $(G,L)$ continuum as $1/(|G|\cdot L)$. The processing gain $10\log_{10}(M)$ dB equals the classical beamforming gain, establishing that this gain is a property of group order, not sensor count. The DFT, DCT, and KLT are unified as group-matched special cases. We conjecture a General Algebraic Averaging Theorem extending these results to arbitrary statistics, with variance governed by the effective group order $d_{\mathrm{eff}}$. Monte Carlo experiments on the first four sample moments across five group types confirm the conjecture to four-digit precision. The framework exploits the $structure$ of information (representation-theoretic symmetry of the data object) rather than the content, complementing Shannon's theory. Five applications are demonstrated: single-snapshot MUSIC, massive MIMO, single-pulse waveform classification, graph signal processing, and analysis of transformer LLMs. Techniques for blind group matching are described.
arXiv:2604.05840v2 Announce Type: replace
Abstract: CHRONOS is a proposed gravitational-wave detector designed to operate in the sub-Hz frequency range (0.1 to 10 Hz), a largely unexplored band due to strong noise sources that hamper ground-based detectors. It employs cryogenic operation, a cross torsion-bar configuration, a triangular Sagnac interferometer, and a speed meter readout scheme to overcome key noise limitations, targeting a strain sensitivity of $h \sim 10^{-18} Hz^{-1/2}$ around 2 Hz and a stochastic gravitational wave background of $\Omega_{GW}$ approximately $2 \times 10^{-3}$ at 2 Hz. Using analytical and interferometric simulations with FINESSE3, we evaluate the noise budget of CHRONOS and characterize the relative contributions of quantum, thermal, and environmental noise sources. Our results demonstrate that CHRONOS achieves competitive sensitivity at low frequencies. The feasibility of using CHRONOS in an earthquake early-warning system by detecting prompt gravity-gradient signals is also investigated, and is predicted to be faster by approximately 2.92 to 6.90 seconds within 40 km. These findings highlight the scientific potential of CHRONOS, bridging gravitational-wave astronomy and geophysical monitoring, and motivating further development of low-frequency detector technologies.
arXiv:2604.15524v3 Announce Type: replace
Abstract: This paper presents a novel density control framework for multi-robot systems with spatial safety and energy sustainability guarantees. Stochastic robot motion is encoded through the Fokker-Planck Partial Differential Equation (PDE) at the density level. Control Lyapunov and control barrier functions are integrated with PDEs to enforce target density tracking, obstacle region avoidance, and energy sufficiency over multiple charging cycles. The resulting quadratic program enables fast in-the-loop implementation that adjusts commands in real-time. Multi-robot experiment and extensive simulations were conducted to demonstrate the effectiveness of the controller under localization and motion uncertainties.
arXiv:2604.23190v2 Announce Type: replace
Abstract: Automating repository-level software engineering tasks is a foundational challenge for autonomous code agents, largely due to the difficulty of configuring executable environments. However, manual configuration remains a labor-intensive bottleneck, necessitating a transition toward fully automated environment configuration. Existing approaches often rely on pre-defined artifacts or are restricted to specific programming languages, limiting their applicability to diverse real-world repositories. In this paper, we first propose RAT (RunAnyThing), a modular and extensible agent framework for fully automated configuration across programming languages on arbitrary repositories. RAT adopts a multi-stage pipeline that integrates language-aware abstraction, image initialization, specialized configuration toolset, and robust sandbox. Furthermore, to enable rigorous evaluation, we propose RATBench, a benchmark reflects the comprehensive coverage of real-world repositories. Extensive experiments demonstrate that RAT achieves state-of-the-art performance, improving Environment Setup Success Rate (ESSR) by an average of 36.1% over strong baselines.
arXiv:2604.23466v2 Announce Type: replace
Abstract: NVIDIA's CUDA Tile (CuTile) introduces a Python-based, tile-centric abstraction for GPU kernel development that aims to simplify programming while retaining Tensor Core and Tensor Memory Accelerator (TMA) efficiency on modern GPUs. We present the first independent, cross-architecture evaluation of CuTile against established approaches such as cuBLAS, Triton, WMMA, and raw SIMT on three NVIDIA GPUs spanning Hopper and Blackwell: H100 NVL, B200, and RTX PRO 6000 Blackwell Server Edition. We benchmark representative AI workloads, including GEMM, fused multi-head attention, and end-to-end LLM inference in BF16/FP16 precision, to assess both performance and portability.
Our results show that CuTile effectiveness is strongly workload- and architecture-dependent. On datacenter-class Blackwell (B200), CuTile achieves up to 1007 TFLOP/s for fused attention, outperforming FlashAttention-2 by 2.5x while requiring only 60 lines of Python kernel code. For GEMM, CuTile reaches 52-79% of cuBLAS performance in 22 lines of code (versus 123 for WMMA), making it a practical replacement for hand-written CUDA kernels but not yet for vendor-optimized libraries. However, the same CuTile attention kernel achieves only 53% of FlashAttention-2 throughput on RTX PRO 6000 (sm_120), exposing significant cross-architecture optimization gaps. In contrast, Triton sustains 62-101% of cuBLAS performance across all tested platforms without architecture-specific tuning, demonstrating substantially stronger portability.
arXiv:2605.08215v2 Announce Type: replace
Abstract: Visual Foresight VLA (VF-VLA) has become a prominent architectural choice in the recent VLA due to its impressive performance. Nevertheless, the inherent design of VF-VLA makes it particularly vulnerable to out-of-distribution (OOD) shifts. Because the quality of action directly depends on the accuracy of the predicted future visual information, OOD conditions affect both stages at once. To address this vulnerability, we propose Test-Time Training Visual Foresight VLA ($T^3$VF), a test-time training approach motivated by the observation that the predicted future image and its subsequent observation form a natural supervision pair. To further address the practical challenges that arise from indiscriminate test-time updates, we introduce an adaptive update filtering mechanism. Empirically, $T^3$VF mitigates the OOD vulnerability of VF-VLA at a modest additional inference cost, without requiring any architectural modification or auxiliary modules.
arXiv:2605.09989v2 Announce Type: replace
Abstract: Recent advances in robot imitation learning have produced powerful visuomotor policies that manipulate diverse objects from visual inputs. However, monocular observations lack depth information, which is critical for precise manipulation in cluttered or geometrically complex scenes. Explicit depth maps and point clouds are often noisy and fragile in real-world manipulation. We introduce StereoPolicy, a visuomotor policy learning framework that directly leverages synchronized stereo image pairs to improve geometric reasoning without constructing explicit 3D representations. StereoPolicy processes each image with pretrained 2D vision encoders and fuses left-right features through a cross-attention-based Stereo Transformer, capturing spatial correspondence and disparity cues implicitly. The framework integrates with diffusion-based and pretrained vision-language-action (VLA) policies, delivering consistent improvements over RGB, RGB-D, point cloud, and multi-view baselines across three simulation benchmarks and seven real-robot tabletop and bimanual mobile manipulation tasks. Our results show that stereo vision bridges 2D pretrained representations and 3D geometric understanding for robotic manipulation.
arXiv:2605.13075v3 Announce Type: replace
Abstract: Few-shot spoken word classification has largely been developed for applications where a small number of classes is considered, and so the potential of larger-scale few-shot spoken word classification remains untapped. This paper investigates the potential of a spoken word classifier to sequentially learn to distinguish between 1000 classes when it is given only five shots per class. We demonstrate that this scaling capability exists by training a model using the Generative Meta-Continual Learning (GeMCL) algorithm and comparing it to repeatedly trained or finetuned baselines. We find that GeMCL produces exceptionally stable performance, and although it does not always outperform a repeatedly fully-finetuned HuBERT model nor a frozen HuBERT model with a repeatedly trained classifier head, it produces comparable performance to the latter while adapting 2000 times faster, having been trained less than half of the data for two orders of magnitude less time.
arXiv:2605.16138v2 Announce Type: replace
Abstract: Neural architecture search (NAS) is a powerful approach for automating model design, but existing methods often optimize for accuracy alone or rely on proxy metrics such as bit operations (BOPs) that correlate poorly with hardware cost. This gap is particularly large for FPGA deployment, where cost is dominated by a multi-dimensional budget of lookup tables, DSPs, flip-flops, BRAM, and latency. We present the Surrogate Neural Architecture Codesign Package (SNAC-Pack), an open-source AutoML framework for hardware-aware neural architecture codesign and end-to-end FPGA deployment. SNAC-Pack runs a multi-objective global search with Optuna and NSGA-II, loading trials to a shared SQLite store that enables parallel workers across compute nodes. A hardware surrogate model outputs per-trial resource and latency estimates, avoiding the synthesis cost that would otherwise dominate the search loop. A local search stage then applies quantization-aware training (QAT) together with iterative magnitude pruning in a combined compression loop, after which the final model is synthesized to FPGA firmware via the hls4ml Python library. A YAML configuration and an optional agentic frontend let users run the pipeline on new datasets without modifying the framework. We demonstrate SNAC-Pack on jet classification at the Large Hadron Collider and superconducting qubit readout, discovering compact architectures that match or exceed strong baselines on the task metric while reducing FPGA resource utilization and, in the qubit readout case, reducing the design space exploration process from months of manual fine-tuning to hours of automated search.