arXiv:2606.06990v1 Announce Type: new
Abstract: The generation of high-fidelity synthetic Electronic Health Records (EHR) is crucial for advancing medical research while preserving patient privacy. However, head-to-head comparison of existing generative models is hindered by disjointed codebases, incompatible data loaders, conflicting library dependencies, and inconsistent evaluation protocols. To address these gaps, we introduce a lightweight, end-to-end benchmarking framework for reproducible synthetic EHR evaluation, organized as a unified pipeline spanning data ingestion, standardized model training, and architecture-agnostic evaluation. Our current implementation targets the generation of longitudinal ICD diagnosis codes -- the most commonly studied modality in this literature -- and is built on the community-maintained PyHealth library. We reimplement and unify strong baselines (MedGAN, CorGAN, PromptEHR, HALO) under full ICD-9 vocabulary granularity, and add a lightweight GPT-2 baseline from the general-purpose sequence-modeling literature. We contribute a rigorous, architecture-agnostic privacy-utility evaluation suite that applies identically to GAN- and transformer-based generators, and report bootstrapped confidence intervals across all metrics. We further analyze the poor long-tailed performance of existing models and discuss the extensibility of our framework beyond diagnosis codes. By lowering the engineering barrier to running, extending, and evaluating under a single pipeline, we introduce a starting point for community-driven reproducibility and benchmarking synthetic EHR models.
Science Journals
arXiv:2606.07453v1 Announce Type: new
Abstract: The Odd Cycle Transversal (OCT) problem, which asks for a minimum subset of vertices whose removal renders a graph bipartite, is a central problem in algorithmic graph theory. It is known to be NP-complete even on $P_k$-free graphs for $k \ge 6$. Furthermore, assuming the Unique Games Conjecture (UGC), OCT does not admit a constant-factor approximation algorithm on general graphs.
Motivated by these hardness results, we investigate the approximability of OCT on $P_k$-free graphs. We first establish that the problem becomes polynomial-time solvable on specific subclasses of $P_k$-free graphs, most notably $(P_6, C_3)$-free graphs, by exploiting a structural decomposition into rings of bipartite graphs. Leveraging these tractable substructures as a basis, we present a constant-factor approximation algorithm for OCT on general $P_k$-free graphs. We achieve an approximation ratio of $k-2$ when $k$ is odd and $k-3$ when $k$ is even. These results provide the first nontrivial constant-factor approximations for this class dependent on $k$, aligning with the UGC implication that no approximation factor independent of $k$ is likely to exist.
arXiv:2606.07249v1 Announce Type: new
Abstract: Accurate monitoring of forest disturbances is essential for understanding carbon dynamics and land management, yet traditional approaches typically rely on pixel-wise analysis of satellite time-series, ignoring spatial context. We present a deep learning framework that maps 38 years (1984-2022) of forest disturbance across the contiguous United States by modeling temporal trajectories and spatial neighborhoods simultaneously. By leveraging a vision transformer architecture, our approach effectively filters noise from weak supervision signals to produce spatially coherent disturbance maps. We perform exhaustive evaluations across multiple satellites (Landsat, Sentinel-1, Sentinel-2) and temporal windows (38 years and the more recent 6 years), validating performance against a novel, manually annotated validation dataset (n=300) and independent fire perimeter dataset (n=706). The results highlight the complexity of the task: while our spatio-temporal model demonstrates high precision (up to 98.2% for +-1 year detection on MTBS and up to 71.3% on the CONUS validation datasets, with F1-scores up to 75.8% and 47.3%, respectively) and effectively reduces spatial artifacts, it exhibits performance trade-offs across different disturbance regimes compared to pixel-wise baselines. Our method offers a promising foundation for consistent forest monitoring.
arXiv:2603.17003v2 Announce Type: replace-cross
Abstract: We propose a constricting Control Barrier Function (CBF) framework for prescribed-time control of control-affine systems with input constraints. Given a system starting outside a target safe set, we construct a time-varying safety tube that shrinks from a relaxed set containing the initial condition to the target set at a user-specified deadline. Any controller rendering this tube forward invariant guarantees prescribed-time recovery by construction. The constriction schedule is bounded and tunable by design, in contrast to prescribed-time methods where control effort diverges near the deadline. Feasibility{\aa} under input constraints reduces to a single verifiable condition on the constriction rate, yielding a closed-form minimum recovery time as a function of control authority and initial violation. The framework imposes a single affine constraint per timestep regardless of state dimension, scaling to settings where grid-based reachability methods are intractable. We validate on an 18-dimensional multi-agent system, demonstrating scalability and prescribed-time recovery with bounded control effort.
arXiv:2606.06711v1 Announce Type: new
Abstract: Synchrotron beamlines differ in hardware, technique, and workflow, making customized control interfaces necessary; bespoke per-beamline graphical user interfaces (GUIs) do not scale well, one-size-fits-all facility software forces compromises that leave most of the interface unused, and even recent component-library approaches keep per-scientist tweaks on a developer's queue. We present Lightfall, a control platform designed for facility-wide use, whose API-first architecture exposes every panel, device, and scan plan through a single uniform addressable interface. An embedded language-model agent drives experiments through that interface, from a single move-and-read to a Gaussian-process-driven autonomous scan, while beamline staff extend the interface during operation via skills: plugin modules the agent invokes to compose and modify panels in the running application. The result is a closed development loop: a beamline scientist authors a panel change in natural language, the agent emits and applies it, and the commit lands in the beamline's plugin repository as a side effect. The per-iteration cost of a scientist-driven change is then fixed in the scientist's own time rather than in developer hours the facility must supply. Lightfall is in testing at the COSMIC-Scattering beamline at the Advanced Light Source.
arXiv:2604.23873v2 Announce Type: replace
Abstract: This paper presents two enhancements to cylindrical algebraic decomposition (CAD) based quantifier elimination (QE) for cases in which multiple equational constraints are present in the given input formula $\phi^*$. The first enhancement provides more detail in the output when there is a conceptual partition of the set of variables of $\phi^*$ into parameters and unknowns. In such cases, we describe how to partition the parameter space so that: (1) in each open set of the partition the number $\nu$ of associated unknowns is a finite constant or is infinite; and (2) for each such open set for which $\nu$ is finite, an expression for the unknowns in terms of the parameters is provided. The second enhancement is an efficiency gain achievable in certain situations. Indeed, when certain conditions are met, the second CAD equational projection step can be reduced more significantly than is supported by the prior existing theory. Relevant theorems and worked examples for both enhancements are provided. Application areas include approximation theory, cuspidal manipulator classification, and biological/chemical systems.
arXiv:2606.07248v1 Announce Type: new
Abstract: Serial LLM inference backends -- such as Ollama -- process requests one at a time under FCFS admission, causing Head-of-Line Blocking (HOLB) under mixed workloads at high utilisation: short factual queries can be delayed by minutes behind long generation jobs. While cloud-scale deployments mitigate HOLB via continuous batching (vLLM, Orca), these solutions require tens of GB of VRAM for concurrent KV-caches -- infeasible for memory-constrained edge and local deployments that rely on serial request dispatch. We present \clairvoyant, a drop-in sidecar proxy for any serial OpenAI-compatible backend (e.g., Ollama, llama.cpp). \clairvoyant predicts response length from 19 lightweight lexical features via an ONNX-exported XGBoost classifier, achieving 0.029\,ms per-request latency (four orders of magnitude below typical generation time). Because admission scheduling depends on relative ordering rather than exact prediction, the system optimises ranking fidelity, achieving 62--96\% in-distribution and 52--66\% cross-distribution accuracy across natural conversation datasets. We find that curated instruction datasets are degenerate training sources for length prediction: GPT-imposed brevity constraints reduce Long-class representation to under 0.02\% of examples, making natural conversation logs the only viable training source. End-to-end GPU benchmarks on an RTX~4090 show 70--76\% P50 latency reduction for short requests under maximum queue pressure (100 concurrent requests) and 17\% under steady-state Poisson arrivals ($\rho=0.74$). \clairvoyant is open-source and requires no modifications to the inference backend.
arXiv:2606.06871v1 Announce Type: new
Abstract: Diagnosing 802.11 packet captures requires expert protocol knowledge, is slow, inconsistent across engineers, and unscalable. LLM-based approaches sound plausible but fabricate protocol events absent from captures (especially truncated traces), produce uncalibrated confidence scores, and suffer evaluation bias when golden references are co-produced by the model under test. We introduce PROBE (Protocol Reasoning Over evidence-Based Ensembles), a multi-stage pipeline addressing all three failures. It integrates (i) deterministic PCAP-to-text normalization with frame-level verifiability, (ii) multi-run, multi-candidate ensembles with optional cross-model second opinion and progressive obfuscation, (iii) a verdict-aware evidence framework treating absence of failure evidence as contributing evidence, and (iv) a fully deterministic composite reliability score from evidence validity, run-to-run stability, and cross-model agreement without LLM self-assessment.
On 87 enterprise Wi-Fi captures (104 capture-reviewer pairs), single-pass LLM analysis raises weighted evidence F1 from 0.871 (expert baseline) to 0.912 but misses critical frames in 35% of cases. Naive ensemble voting drops below baseline (0.842) as majority voting amplifies conservative verdicts: 50% of confirmed failures are misclassified as 'no issue' or 'insufficient evidence.' Adding evidence-grounded reconciliation achieves 0.957 F1, a 96% auto-accept rate, and a worst-case floor above 0.70. LLM self-reported confidence clusters at 0.95 regardless of difficulty (71% report exactly 0.95), confirming it is uninformative. We also introduce a model-agnostic evaluation framework using per-field assertion matching, eliminating circular bias from model-co-produced golden references.
arXiv:2606.06862v1 Announce Type: new
Abstract: Accurate and real-time monitoring of spatial beam quality has emerged as the absolute prerequisite for intelligent optical field regulation and advanced laser applications. However, modern high-power and multimode optical systems exhibit highly complex, nonlinear, and transient behaviors. In these systems, the spatial beam profile undergoes dramatic reorganizations within extremely short timeframes. Phenomena such as spatio-temporal mode-locking, transient beam self-cleaning, and plasma-induced aberrations demand nanosecond-level dynamic characterization. Yet, capturing these ultrafast dynamics is fundamentally bottlenecked by the kilohertz frame rates of conventional two-dimensional image sensors. To break this dimensional and temporal barrier, we propose an ultrafast non-imaging beam quality monitoring technique, termed Fiber-based Laser Assessment via Spatial-to-temporal High-speed-mapping (FLASH). By utilizing a multimode fiber to encode spatial beam variations into high-dimensional speckle fingerprints and a multicore fiber delay line array to serialize these features, we transform two-dimensional spatial information into high-speed one-dimensional temporal pulse sequences. Empowered by a deep learning model to decipher the serialized signals, the FLASH system achieves an unprecedented 100 MHz measurement rate with a minimal mean relative error of 0.32%. Realizing a five-order-of-magnitude speed improvement over standard camera-based methods, this spatial-to-temporal mapping paradigm provides a transformative spatial oscilloscope. It unlocks new possibilities for real-time intelligent adaptive control and the exploration of complex multimode nonlinear physics.
arXiv:2606.06721v1 Announce Type: new
Abstract: Robots that operate over extended periods should not merely visit space; they should progressively understand it. Yet most 3D scene graph pipelines treat perception as a post-processing stage over a fixed dataset, decoupling scene representation from the decisions that determine what is observed in the first place. We present SCOUT, an online semantic exploration framework that closes this loop by coupling active traversal with probabilistic scene graph construction. Given a prior 2D occupancy map and posed RGB-D observations, SCOUT incrementally builds an uncertainty-aware 3D scene graph whose nodes maintain fused geometry and posterior beliefs over open-vocabulary object labels, while edges encode structural relations such as on, inside, belong, and next to. These beliefs are fed back to an uncertainty-guided traversal planner, which selects viewpoints by balancing expected semantic certainty gain, geometric coverage gain, and travel cost. In this way, the robot revisits ambiguous objects when additional evidence matters and expands into unseen free space when the scene remains incomplete. The resulting system treats semantic scene completeness as an operational objective rather than a passive by-product of semantic mapping, moving toward autonomous agents that can patrol, update, and reason about evolving indoor environments with minimal human intervention.
arXiv:2606.06726v1 Announce Type: new
Abstract: Configuring network access control policies in large, complex networks is error-prone and requires significant expert effort. LLMs offer a promising interface for expressing such policies in natural language, but their capability for translating user requests into access policies, and the system architectures best suited to leverage LLMs, remain underexplored. We present an architecture for natural-language access control (NLAC) that uses LLMs to translate user requests into access policies, and introduce NLACBench, a benchmark for evaluating LLM-based intent translation systems in large-scale networks. Our evaluation across multiple state-of-the-art models shows that top-performing LLMs achieve up to 96.9% accuracy in small-network settings, but performance degrades substantially (below 20% for some models) as network size increases. To address this limitation, we identify relevant network components via embedding similarity and construct compact subgraphs that are passed to the LLM. This approach enables scaling to larger networks with up to 98.7% accuracy, while simultaneously reducing inference time, hardware requirements, and operating costs to a constant resource budget. Finally, a case study indicates that top-performing models exhibit largely complementary error patterns, suggesting that intent translation accuracy may be further improved through multi-LLM architectures.
arXiv:2604.17433v2 Announce Type: replace
Abstract: Self-consistency (SC) is a popular technique for improving the reasoning accuracy of large language models by aggregating multiple sampled outputs, but it comes at a high computational cost due to extensive sampling. We introduce a hybrid ensembling approach that leverages the complementary strengths of two distinct modes of reasoning: Chain-of-Thought (CoT) and Program-of-Thought (PoT). We describe a general framework for combining these two forms of reasoning in self-consistency, as well as particular strategies for both full sampling and early-stopping. We show that CoT-PoT ensembling not only improves overall accuracy, but also drastically reduces the number of samples required for SC by a factor of 9.3x. In particular, the majority of tasks (78.6%) can be addressed with only two samples, which has not been possible with any prior SC methods.
arXiv:2604.18401v3 Announce Type: replace
Abstract: Agentic reinforcement learning (RL) is emerging as a critical post-training paradigm for improving LLM agent capabilities. Existing RL algorithms for LLMs largely follow the token-centric paradigm as in RLHF and RLVR, where tokens serve as the basic units for modeling and optimization. However, this paradigm introduces a granularity mismatch in agentic RL, as it optimizes token-level predictions while LLM agents make step-level decisions through cycles of environmental observations and actions. To bridge this gap, we propose \textbf{StepPO}, a step-centric paradigm for agentic RL via step-aligned policy optimization. Specifically, we reformulate agentic RL from a token-level Markov Decision Process (MDP) into a step-level MDP, where interaction steps serve as the basic trajectory representations. We further propose step-level credit assignment to align policy optimization with the natural granularity of agent decisions. Together, StepPO optimizes agent policies at the step level for multi-turn agent-environment interaction. Experiments across multi-hop QA, academic paper search, and text-world action tasks show that StepPO consistently outperforms various RL algorithms. Further analyses provide insights into how step-centric paradigm improves agent training. We hope this step-centric paradigm offers a useful lens for understanding agent behavior and a practical path for training more capable LLM agents.
arXiv:2605.01642v2 Announce Type: replace
Abstract: Prevailing alignment methods target a fixed set of preferences and therefore risk forcing value lock-in as societal norms evolve over time. We introduce Adaptive Pluralistic Alignment (APA), a modular pipeline for updating pluralistically aligned AI systems to track evolving values and avoid value lock-in without repeating costly pretraining or large-scale data collection. APA has three stages: (1) learning compact personalized reward models via low-rank reward basis decomposition, (2) using these models as a jury that collectively selects among candidate outputs through social-choice-theoretic voting, and (3) efficiently adapting the jury over time by fitting new annotator weights over the fixed reward bases as values shift. The resulting system is efficient, explainable, steerable, and modular. We implement a proof-of-concept instantiation using the PRISM multi-user alignment dataset and simulated historical annotators, and provide preliminary analysis showing that jury composition and the choice of voting rule can substantially affect outcomes, particularly when jury preferences are heterogeneous. We provide full code and resulting preference datasets at https://github.com/RachelFreedman/apa.
arXiv:2602.16908v2 Announce Type: replace-cross
Abstract: Allegro is a machine learning interatomic potential model designed to predict atomic properties in molecules using E(3) equivariant neural networks. When training this model, there tends to be a trade-off between accuracy and inference time. For this reason, we apply multi-objective hyperparameter optimization to both objectives. Additionally, we experiment with modified architectures by constructing variants of Allegro: one extended with additional classical layers and one incorporating quantum-classical hybrid layers. We evaluate all models on QM9, rMD17-aspirin, rMD17-benzene, and a self-generated dataset of copper-lithium structures. As results, both variants surpass Allegro in force prediction accuracy across multiple datasets. The classical variant consistently improves over the baseline, while the quantum-classical hybrid variant achieves the best overall force prediction accuracy on the Cu-Li dataset, where it was fully optimized, outperforming the classical variant by approximately 13%. Notably, the hybrid variant also achieves competitive results on the remaining datasets despite using hyperparameters transferred from Cu-Li without dataset-specific optimization, suggesting that quantum-classical hybridization is a promising direction for enhancing MLIP architectures.
arXiv:2605.20950v2 Announce Type: replace
Abstract: Vision-Language Models (VLMs) face a bottleneck of prohibitive computational costs arising from massive visual token sequences during inference. Existing vision token reduction methods alleviate this burden, but they unintentionally preserve the isolated visual subject strictly aligned with the user's query, which fails to substantially explore salient subjects and their contextual relationships. In this paper, we propose SPpruner, a subject-centric progressive reduction paradigm that emulates the \textit{Focus-then-Context} mechanism of the human visual perception system. Specifically, we first construct a focus identification module to explicitly model the interplay between visual saliency and semantic relevance. Herein, it can excavate the comprehensive visual subject spectrum to ensure a high-fidelity representation of visual input. Subsequently, a context-aware structural scanning module is developed to aggregate contextual cues from neighboring regions. As such, it can effectively restore global relational dependencies to uphold the structural integrity of the preserved subjects. Extensive experiments demonstrate that our paradigm consistently outperforms SOTA methods, achieving up to 2.53 times speedup with only 22.2% of visual tokens retained in Qwen2.5-VL and a 67% FLOPs reduction on LLaVA with a negligible 0.6% accuracy drop.
arXiv:2606.06861v1 Announce Type: new
Abstract: Understanding nonlinear feature interactions is crucial in science and engineering, yet standard multilayer perceptrons (MLPs) often capture such interactions only implicitly, leading to entangled representations that can impair robustness and interpretability. We investigate product-unit residual networks (PURe) that integrate multiplicative product units with residual connections to explicitly model cross-feature couplings while stabilizing optimization. We conduct a systematic evaluation on an interaction-driven synthetic benchmark and two real-world datasets, assessing predictive accuracy, robustness to Gaussian feature noise, and performance under limited training data, and we compare real- and complex-valued variants under a matched parameter budget. Beyond accuracy, SHapley Additive exPlanations (SHAP)-based interaction analyses show that PURe learns more concentrated and structurally coherent interaction patterns than MLP baselines. Overall, PURe achieves competitive or improved performance, better robustness and sample efficiency in low-data regimes, and enhanced interaction-level interpretability.
arXiv:2606.06559v1 Announce Type: new
Abstract: Full-duplex spoken dialogue models allow voice agents to listen and speak concurrently, enabling natural interaction with real-time overlap. However, end-to-end dual-channel models that jointly encode user and agent streams may degrade in realistic acoustic environments: interfering speakers leaking into the user microphone can be encoded as part of the user query, corrupting the LLM's conditioning and causing unstable turn-taking and reduced response quality. We propose Interference-Resilient Adaptive Fusion (IRAF), a lightweight, streaming-compatible module that modulates the contribution of user audio to the LLM frame by frame. IRAF predicts a scalar reliability gate from target-speaker and user audio embeddings and rescales user representations before fusion with agent embeddings. Experiments on MS-MARCO and InstructS2S-200K show consistent gains in response quality and full-duplex interaction under interfering-speaker conditions.
arXiv:2605.06647v2 Announce Type: replace
Abstract: Retrieval-augmented agents are increasingly the interface to large knowledge bases, yet most treat retrieval as a black box: they issue exploratory queries, inspect snippets, and reformulate until evidence emerges. This resembles how a newcomer searches an unfamiliar database rather than how an expert navigates it with strong priors about terminology and likely evidence, causing extra retrieval rounds, latency, and poor recall.
We introduce \textit{Superintelligent Retrieval Agent} (SIRA), which casts \emph{superintelligence} in retrieval as compressing multi-round exploratory search into a single corpus-discriminative retrieval action. SIRA does not merely ask which terms are relevant; it asks which terms separate the desired evidence from corpus-level confusers. Offline, an LLM enriches each document with missing search vocabulary; at query time, it predicts evidence vocabulary the query omits; and corpus statistics serve as tool calls that filter terms that are absent, overly common, or unlikely to create retrieval margin. The final step is a single weighted BM25 call combining the query with the validated expansion.
Across ten BEIR benchmarks, SIRA achieves the strongest average retrieval performance in our comparison, beating dense retrievers, learned sparse retrievers, and LLM search-agent baselines while using no relevance labels or retriever fine-tuning. On downstream QA, its retrieval-only answer coverage exceeds recent RL-trained agentic QA systems on NQ and HotpotQA. We also introduce \textbf{BrowseComp-Wikipedia}, a hard-search benchmark of 232 BrowseComp-derived queries over a 25,587,229-document Wikipedia index. Even without index-time enrichment, using only grounded Wikipedia categories, SIRA outperforms multi-round Perplexity agents at every budget, reaching 9.70% Recall@1, 15.27% Recall@10, and 36.14% Recall@100.
arXiv:2606.06572v1 Announce Type: new
Abstract: We argue that modern generative models create structural risks for knowledge and cultural production at current, sub-AGI capability levels. We define Human Temporal Learning (HTL) as path-dependent knowledge accumulation through sustained engagement with problems over time. Generative outputs increasingly resemble HTL-intensive work in surface features, so verifying whether a given output reflects genuine human learning grows costly relative to its expected benefit. Once verification loses economic justification, evaluators reward outputs regardless of production mode, and producers who invested years of learning compete on price against outputs that cost almost nothing to generate. We call this pathway value collapse and formalize it through a costly-inspection framework. Cross-domain evidence from academic publishing, legal practice, content platforms, and software security maps onto four stages of verification erosion. Alignment success is orthogonal. Better-aligned models narrow observable gaps between human and AI outputs, making source verification harder and intensifying competitive pressure against HTL-intensive work even when individual AI outputs improve.
arXiv:2606.06851v1 Announce Type: new
Abstract: The Learning Analytics (LA) community has undergone rapid development over the 15 years since the first LAK conference was held. However, while epistemological and ethical debates regarding the philosophical foundations of LA have been vigorous, metaphysical discussions have been sparse, signifying a lack of effort to derive the identity of LA from its internal principles. In this paper, we attempt to establish a metaphysics of LA by addressing the ontological question of ``What is LA?'' We do so by tracing back to LA's own definitions and principles to derive an answer from within LA itself. Specifically, we address what kind of existence the data LA operates on constitutes, identify eight agents including learners as ontological prerequisites, and clarify, via the is/ought problem, that LA does not derive norms from data. In particular, this system reveals that a class of LA practices, here termed \textit{norm-embedded LA}, conflates LA's purpose with its operations, creating an ontological tension with the first principle. We also discuss connections with related fields and the limitations of this system. The metaphysics outlined here is not imposed from outside LA, but surfaces what LA itself has always implicitly presupposed.
arXiv:2601.10896v2 Announce Type: replace
Abstract: LLMs are increasingly used as third-party judges, yet their reliability when evaluating speakers in dialogue remains poorly understood. We show that LLMs judge identical claims differently depending on framing: the same content receives different verdicts when presented as a statement to verify ("Is this statement correct?") versus attributed to a speaker ("Is this speaker correct?"). We call this dialogic deference and introduce DialDefer, a framework for detecting and mitigating these framing-induced judgment shifts. Our Dialogic Deference Score (DDS) captures directional shifts that aggregate accuracy obscures. Across ten domains, 3k+ instances, and five models, conversational framing induces large shifts (mean|DDS|=15.9 percentage points (pp) across models, p < .0001) while accuracy remains stable (<2 pp), with effects amplifying 2--5x on naturalistic Reddit conversations. This effect is domain-dependent: a single model can shift toward disagreement (skepticism) on graduate-level science and toward agreement (deference) on social judgment. Ablations reveal that human-vs-LLM attribution drives the largest shifts (17.7 pp swing), suggesting models treat disagreement with humans as more costly than with AI. Mitigation attempts can reduce deference but over-correct into skepticism, revealing a calibration problem beyond accuracy optimization.
arXiv:2606.06528v1 Announce Type: new
Abstract: In resource-constrained small-satellite settings, AI inference must operate under tight size, power, and payload budgets, which tend to limit onboard compute capability and data handling. These conditions motivate establishing a clear baseline for quantized AI inference under bounded compute and memory resources. To instantiate this baseline, a representative embedded-vision neural-network workload serves as the reference case. With this motivation, this paper presents a measurement-based characterization of quantized execution for this AI workload on highly constrained embedded platforms (for instance, Cortex-M), grounded as a lower-bound operating point. In this regime, scaling tends to rely on explicit orchestration rather than OS-managed, transparent multicore scheduling, and timing behavior is shaped by instruction efficiency and memory movement. As a result, the characterization provides a structured reference for estimating execution time across orchestrated configurations (e.g., multiple cores and/or devices), treating orchestration and architectural variation as explicit design choices. We report latency metrics alongside data-movement observations, and interpret these measurements in light of ALU/SIMD utilization under quantized arithmetic for the Cortex-M. Finally, we outline how this baseline provides a reference point for positioning the results against more space-typical embedded processor classes (e.g., LEON/NOEL-V).
arXiv:2606.07151v1 Announce Type: new
Abstract: Traditional change point detection in dynamic networks assumes abrupt transitions between stationary states, overlooking scenarios of continuous evolution which arise in most real-world applications, such as social networks or physical systems. We address this gap by formally defining regimes as periods of coherent dynamics in temporal graphs, which we characterize as trajectories along geodesics in a suitably defined graph space. This original perspective allows us to define regime changes as significant drifts in dynamics, either toward new trajectories or with pace changes. We leverage graph regression methods to measure the cumulative distance of sequences of observed graphs from the estimated geodesics between their endpoints, in the relevant graph space, which we can combine with change point detection algorithms. We present experiments on dynamic networks, with changing trajectories and varying speeds, in which we outperform state of the art change point detection models. Then, we analyse mobility data during the Covid-19 pandemic, and show that our assumptions on regular network evolution lead to change points that are more aligned to external events compared to the outcomes of baseline methods. Our work is the first to model and detect changes between evolving regimes in graph space, providing a realistic and powerful tool for analyzing complex temporal graph data.
arXiv:2606.07153v1 Announce Type: new
Abstract: Physics-informed learning is increasingly used for partial differential equation (PDE)-governed inverse problems, but its reliability remains difficult to certify. This paper develops a no-harm certification-and-selection framework for physics-informed inverse learning. A learned reconstruction is accepted only when its residual-calibrated radius is no worse than the baseline radius, namely when $$R_{\mathrm{learn}}\le R_{\mathrm{base}}+\varepsilon_{\mathrm{safe}};$$otherwise, the method returns the baseline. The certificate combines data, physics, boundary or initial-condition, and optimization residuals. Under a conditional stability estimate, these residuals yield an a posteriori reconstruction-error bound and a deterministic uncertainty radius. A high-probability certificate is also derived for physics residuals estimated from independent random collocation points. Numerical tests on Poisson source recovery, inverse heat reconstruction, limited-angle tomography, elliptic coefficient identification, and stochastic residual validation show that the selector accepts certified improvements, rejects shifted, hallucinated, or unfinished candidates, and becomes conservative in strongly ill-posed regimes. The framework is therefore a certification-and-selection layer, not another reconstruction architecture.