arXiv:2605.09817v2 Announce Type: replace Abstract: Agent tools are becoming a core interface through which LLM agents access external data, services, and execution environments. As these tools are distributed through public marketplaces, raw tool counts may substantially overstate ecosystem diversity if many repositories are cloned, lightly modified, or derived from shared templates. Such hidden duplication can contaminate benchmark splits, propagate vulnerable implementations, bias measurements of tool-use generalization, and raise provenance, attribution, and intellectual-property concerns. We present, to our knowledge, the first large-scale measurement study of tool cloning in agentic AI ecosystems. We curate a unified dataset from multiple public platforms, covering 7,508 Model Context Protocol (MCP) repositories with 87,564 extracted tools and 1,353 Skills repositories with 12,447 tools, for a total of 8,861 repositories and 100,011 tool entries. To measure implementation-level duplication, we build a repository-level auditing pipeline using complementary lexical and fuzzy-structural similarity metrics, and compute pairwise similarity across MCP-to-MCP, Skills-to-Skills, and MCP-to-Skills repository pairs. We further manually verify 100 sampled pairs per MCP and Skills ecosystem across similarity-score buckets to calibrate how often high similarity reflects true code cloning. Our analysis shows that cloning is not an isolated artifact: high-similarity regions appear across comparison settings, and 60\% of high-Jaccard candidates and 85\% of high-ssdeep candidates in the MCP ecosystem are manually verified as clones. These results indicate that tool cloning is a pervasive and severe source of hidden duplication in agent-tool ecosystems. They further suggest that agent-tool datasets and benchmarks should account for repository provenance and implementation similarity when measuring tool diversity or constructing evaluation splits.
Science Journals
arXiv:2605.18720v1 Announce Type: new Abstract: Developing dynamic models for tendon-driven continuum robots is challenging due to their nonlinear, high-dimensional, and friction-dominated dynamics. This paper presents a comparative study of data-driven system identification methods, including N4SID, ARX, and SINDYc, for modeling a tendon-actuated continuum robot with rolling joints developed at CERN. Despite the high number of joints of the robot, experimental analysis reveals that a two-degree-of-freedom dynamic model can accurately capture the system dynamics, owing to strong kinematic dependencies between the joints. The models are validated against experimental data, and used in the design of a model predictive controller, demonstrating their feasibility for real-time control.
arXiv:2605.17660v1 Announce Type: cross Abstract: Transformers have become the dominant architecture in modern machine learning, yet the theoretical understanding of their training dynamics remains limited. This paper develops a rigorous mathematical framework for analyzing gradient-based training of transformers in the mean-field regime, where both the depth (number of layers) and width (number of attention heads) tend to infinity. While ResNet training can be understood as controlling a neural ODE, transformer training corresponds to controlling a neural PDE, due to the coupling of multiple token distributions through the attention mechanism. Our mean-field model features two types of measure representations: token distributions evolving through layers and attention parameters at each layer. We establish well-posedness of the forward pass through infinitely deep transformers, characterizing token evolution via flow maps that satisfy ODEs in function spaces. Using adjoint sensitivity analysis, we derive an explicit formula for the conditional Wasserstein gradient of the training risk, involving adjoint variables governed by backward ODEs. We prove the existence and uniqueness of gradient flow curves in the conditional Wasserstein metric space, establishing a rigorous foundation for gradient-based transformer training. A key technical contribution is providing necessary and sufficient conditions for injectivity of the Neural Tangent Kernel (NTK) for attention mechanisms: we show that NTK injectivity is equivalent to linear independence of log-sum-exp functions modulo affine functions, a condition satisfied by diverse token distributions, including discrete distributions, uniform distributions, and Gaussian mixtures. Under this NTK injectivity assumption, we prove that gradient flow converges to global minima when the initial loss is sufficiently small, eliminating spurious local minima from the optimization landscape.
arXiv:2505.18991v3 Announce Type: replace Abstract: Pansharpening seeks to fuse high-resolution panchromatic (PAN) and low-resolution multispectral (LRMS) images into a single image with both fine spatial and rich spectral detail. Despite progress in deep learning-based approaches, existing methods often fail to capture global priors inherent in remote sensing data distributions. Diffusion-based models have recently emerged as promising solutions due to their powerful distribution mapping capabilities, however, they suffer from heavy inference latency. We introduce KSDiff, a fast kernel-space diffusion framework that generates convolutional kernels enriched with global context to enhance pansharpening quality and accelerate inference. Specifically, KSDiff constructs these kernels through the integration of a low-rank core tensor generator and a unified factor generator, orchestrated by a structure-aware multi-head attention mechanism. We further introduce a two-stage training strategy tailored for pansharpening, facilitating integration into existing pansharpening architectures. Experiments show that KSDiff achieves superior performance compared to recent promising methods, and with over $500 \times$ faster inference than diffusion-based pansharpening baselines. Ablation studies, visualizations and further evaluations substantiate the effectiveness of our approach. Code will be released upon possible acceptance.
arXiv:2605.17282v1 Announce Type: cross Abstract: Estimating the largest Lyapunov exponent from a scalar time series is difficult when the governing equations, tangent dynamics, and full state vector are unavailable. We propose FEG-Pro, a forecast-error growth profiling framework for nonlinear scalar time series. The method constructs autocorrelation-guided sparse histories, performs distance-weighted k-nearest-neighbor multi-horizon forecasting, and analyzes the logarithmic growth of geometrically averaged forecast errors. Its primary output is the finite-horizon forecast-error growth slope, lambda_FEG. When the error-growth curve supports a quasi-linear regime, this slope can be compared with reference largest Lyapunov exponents as an estimate of the dominant instability rate. The same pipeline also extracts the formal fit-selection regime, curvature, residual roughness after quadratic detrending, monotonicity, and forecast-error distribution entropy (FEDE) from signed multi-horizon errors. These secondary descriptors are intended not only as diagnostic controls for the slope, but also as candidate machine-learning features for nonlinear signal analysis, because they encode profile geometry and distributional uncertainty not captured by lambda_FEG alone. We evaluate the method on chaotic maps, Mackey-Glass delay dynamics, and scalar Lorenz-63 observables with known or reference exponents. Full-record experiments show good agreement in quasi-linear cases and meaningful curve-shape information in curved or weak profiles. A dyadic length-halving experiment on representative logistic, Mackey-Glass, and Lorenz records shows that residual roughness and mean FEDE often change monotonically and remain interpretable as record length decreases, even when the slope becomes biased or highly variable. The results support treating forecast-error growth as a structured profile and feature-generation framework rather than a single-number estimator.
arXiv:2602.11699v3 Announce Type: replace Abstract: Nonsensical and anomalous sentences have been instrumental in the development of computational models of semantic interpretation. A core challenge is to distinguish between what is merely anomalous (but can be interpreted given a supporting context) and what is truly nonsensical. However, it is unclear (a) how nonsensical, rather than merely anomalous, existing datasets are; and (b) how well LLMs can make this distinction. In this paper, we answer both questions by collecting sensicality judgments from human raters and LLMs on sentences from five semantically deviant datasets: both context-free and when providing a context. We find that raters consider most sentences at most anomalous, and only a few as properly nonsensical. We also show that LLMs are substantially skilled in generating plausible contexts for anomalous cases.
arXiv:2602.16473v2 Announce Type: replace Abstract: C-RASP is a simple programming language that was recently shown to capture concepts expressible by transformers. In this paper, we develop new algorithmic techniques for automatically verifying C-RASPs. To this end, we establish a connection to the verification of synchronous dataflow programs in Lustre, which enables us to exploit state-of-the-art model checkers utilizing highly optimized SMT-solvers. Our second contribution addresses learning a C-RASP program in the first place. To this end, we provide a new algorithm for learning a C-RASP from examples using local search. We demonstrate efficacy of our implementation for benchmarks of C-RASPs in the literature, in particular in connection to the following applications: (1) transformer program optimization, and (2) constrained learning of transformer programs (based on a partial specification).
arXiv:2605.13161v2 Announce Type: replace Abstract: Efficient transfer learning methods for large-scale vision-language models ($e.g.$, CLIP) enable strong few-shot transfer, yet existing adaptation methods follow a fixed fine-tuning paradigm that implicitly assumes a uniform importance of the image and text branches, which has not been systematically studied in image classification. Through extensive analysis, we reveal a Branch Bias issue in vision-language image classification: adapting the image encoder does not always improve performance under out-of-distribution settings. Motivated by this observation, we propose A$_3$B$_2$, an Adaptive Asymmetric Adapter that alleviates Branch Bias in few-shot learning. A$_3$B$_2$ introduces Uncertainty-Aware Adapter Dampening (UAAD), which automatically suppresses image-branch adaptation when prediction uncertainty is high, enabling soft and data-driven control without manual intervention. Architecturally, A$_3$B$_2$ adopts a lightweight asymmetric design inspired by mixture-of-experts with Load Balancing Regularization. Extensive experiments on three few-shot image classification tasks across 11 datasets demonstrate that A$_3$B$_2$ consistently outperforms 11 competitive prompt- and adapter-based baselines.
arXiv:2603.15460v2 Announce Type: replace Abstract: The construction of trial wave functions based on neural networks combined with the variational Monte Carlo method is discussed. The mathematical formulation for representing quantum states as artificial neural networks is introduced. The advantages of employing such trial states and how machine learning works are discussed. It is shown that the variational method is a kind of unsupervised learning algorithm, where the multiple minima landscape is used as an asset that leads to a stable optimization procedure. The feature representation plays an important role on interpretability and on extracting physical insights from nontrivial trial wave functions. The algorithm is illustrated for the Yukawa potential and the hydrogen molecule.
arXiv:2603.25723v2 Announce Type: replace Abstract: Agent performance is strongly shaped by the surrounding harness: the external execution system around a model that organizes a task run. Yet this logic is usually buried in tightly coupled controller code, which makes harnesses hard to inspect, compare, transfer, and ablate. This paper asks whether the reusable design pattern of an agent harness can be represented as an executable natural-language object. We introduce Natural-Language Agent Harnesses (NLAHs), editable documents that describe run-level harness policy, and Intelligent Harness Runtime (IHR), a shared runtime that interprets these documents into agent calls, handoffs, state updates, validation gates, and artifact contracts. Across coding, terminal-use, and computer-use benchmarks, IHR-executed NLAHs achieve comparable task outcomes to code and prompted realizations, while exposing much shorter static harness policies. Module ablations further show that explicit harness modules are analyzable. These results suggest that agent harnesses can be turned from incidental glue around models into scientific representation objects.
arXiv:2604.06579v2 Announce Type: replace Abstract: The Ethereum state database uses Merkle Patricia Trie (MPT), which suffers from large witness proof sizes and high storage overhead. Verkle Tries have been proposed as a replacement, offering witness proofs below 150 bytes through vector commitments and Inner Product Argument aggregation. However, deploying a Verkle Trie in a high-throughput, short block-time blockchain such as Sonic, which produces a block every 300 milliseconds, introduces substantial engineering challenges related to storage efficiency, commitment computation costs, and the need to serve both live and historical state queries in real time. We present SonicDB S6, a production-grade Rust Verkle Trie database for the Sonic blockchain, which leverages its non-forking property to enable aggressive storage optimizations. Occupancy-aware node specializations, selected via an $\mathcal{O}(k n^2)$ dynamic program, reduce live storage by 97.8\%. Delta nodes that record only changed slots reduce archive storage by 95\%. Batched updates, multi-threaded commitment computation, and homomorphic Pedersen caching yield $3.2\times$ higher throughput than a persistent Geth Verkle baseline while sustaining production block-rate performance.
arXiv:2604.23538v3 Announce Type: replace Abstract: In the digital era, personal data, particularly sensitive identifiers such as the Social Security Number and National Identification Number, has become a highly valuable asset, raising significant concerns regarding privacy and security. This study examines the risks associated with the online exposure of the Thai National Identification Number, a key element of identity verification in both governmental and commercial transactions. Similar to the Social Security Number in the United States, this unique identifier is crucial for various legal, financial, and welfare-related activities. However, the increasing digitization of personal records has heightened its vulnerability to unauthorized access and misuse, particularly through search engines that inadvertently index sensitive information. This research identifies publicly exposed Thai National Identification Numbers across major search engines, assessing the potential threats to individual privacy and national security. The study reveals the exposure of over 1.2 million unique National Identification Numbers, along with other highly sensitive personal data, e.g., addresses, contact details, employment status, disability status, and health information. Notably, the analysis indicates that a significant majority of these exposures originate from the Thai government sector websites, highlighting critical vulnerabilities in public data management practices. This widespread exposure not only increases the risk of identity theft and financial fraud but also underscores the urgent need for enhanced cybersecurity measures, stricter regulatory enforcement, and improved data governance within government agencies to prevent future breaches. Addressing these issues is essential to safeguarding citizens' personal information and ensuring compliance with Thailand's data protection laws in an increasingly digitized world.
arXiv:2605.00264v2 Announce Type: replace Abstract: Offline multi-agent reinforcement learning in general-sum settings is challenged by the distribution shift between logged datasets and target equilibrium policies. While standard methods rely on manual pessimistic penalties, we demonstrate that KL regularization suffices to stabilize learning and achieve equilibrium recovery. We propose General-sum Anchored Nash Equilibrium (GANE), which recovers regularized Nash equilibria at an accelerated statistical rate of $\widetilde{O}(1/n)$. For computational tractability, we develop General-sum Anchored Mirror Descent (GAMD), an iterative algorithm converging to a Coarse Correlated Equilibrium at the standard rate of $\widetilde{O}(1/\sqrt{n}+1/T)$. These results establish KL regularization as a standalone mechanism for pessimism-free offline learning that achieves equivalent or accelerated rates in multi-player general-sum games.
arXiv:2605.14133v2 Announce Type: replace Abstract: Interactive agent benchmarks face a tension between scalable construction and realistic workflow evaluation. Hand-authored tasks are expensive to extend and revise, while static prompt evaluation misses failures that only appear when agents operate over persistent state. Existing interactive benchmarks have advanced agent evaluation significantly, but most initialize tasks from clean state and do not systematically test how agents handle pre-existing partial, stale, or conflicting artifacts. We present \textbf{ClawForge}, a generator-backed benchmark framework for executable command-line workflows under state conflict. The framework compiles scenario templates, grounded slots, initialized state, reference trajectories, and validators into reproducible task specifications, and evaluates agents step by step over persistent workflow surfaces using normalized end state and observable side effects rather than exact trajectory matching. We instantiate this framework as the ClawForge-Bench (17 scenarios, 6 ability categories). Results across seven frontier models show that the best model reaches only 45.3% strict accuracy, wrong-state replacement remains below 17\% for all models, and the widest model separation (17% to 90%) is driven by whether agents inspect existing state before acting. Partial-credit and step-efficiency analyses further reveal that many failures are near-miss closures rather than early breakdowns, and that models exhibit qualitatively different failure styles under state conflict.
arXiv:2605.14381v2 Announce Type: replace Abstract: Recent advancements in generative AI facilitate large-scale synthetic data generation for model evaluation. However, without targeted approaches, these datasets often lack the sociotechnical nuance required for sensitive domains. We introduce NodeSynth, an evidence-grounded methodology that generates socially relevant synthetic queries by leveraging a fine-tuned taxonomy generator (TaG) anchored in real-world evidence. Evaluated against four mainstream LLMs (e.g., Claude 4.5 Haiku), NodeSynth elicited failure rates up to five times higher than human-authored benchmarks. Ablation studies confirm that our granular taxonomic expansion significantly drives these failure rates, while independent validation reveals critical deficiencies in prominent guard models (e.g., Llama-Guard-3). We open-source our end-to-end research prototype and datasets to enable scalable, high-stakes model evaluation and targeted safety interventions (https://github.com/google-research/nodesynth).
arXiv:2605.17830v1 Announce Type: new Abstract: Safety evaluations of memory-equipped LLM agents typically measure within-task safety: whether an agent completes a single scenario safely, often under adversarial conditions such as prompt injection or memory poisoning. In deployment, however, a single agent serves many independent tasks over a long horizon, and memory accumulated during earlier tasks can affect behavior on later, unrelated ones. Studying this regime requires evaluation along the temporal dimension across tasks: not whether an agent is safe at any single memory state, but how its safety profile changes as memory accumulates across many independent interactions. We call this failure mode temporal memory contamination. To isolate memory exposure from stream non-stationarity, we introduce a trigger-probe protocol that evaluates a fixed probe set against read-only memory snapshots at varying prefix lengths, together with a NullMemory counterfactual baseline for identifying memory-induced violations. We apply this protocol across three deployment scenarios spanning records, memos, forms, and email correspondence and eight memory architectures, and additionally on Claw-like AI agents, such as OpenClaw, using the platform's native memory mechanism. Memory-enabled agents consistently exceed the NullMemory baseline, and memory-induced violation rates show a robust upward trend with exposure length on both agent classes. Order-randomization experiments indicate that the effect is driven primarily by accumulated content rather than encounter order. Finally, a structural consequence of the event decomposition is that memory-induced risk is detectable from retrieval state before generation, which we confirm with a high-recall diagnostic monitor. Our results argue for treating memory safety as a longitudinal property that requires temporal evaluation, not a single-state property that can be captured by a snapshot.
arXiv:2605.17829v1 Announce Type: new Abstract: AI evaluation is undergoing a structural change. Large language models (LLMs) are increasingly deployed as systems that act over time through tools, environments, users, and other agents, while many evaluation practices still inherit assumptions from response-centered benchmarks (e.g., fixed inputs, isolated outputs, and outcome judgments that can be made from a single response). The field has begun to build interactive benchmarks, but the resulting landscape is fragmented: benchmarks differ in what interaction artifacts they admit, how trajectories are scored, and what claims their results support. This position paper argues that interactive evaluation should be treated as a principled evaluation paradigm, not merely a new family of agent benchmarks. Simply adopting previous evaluation paradigms does not suffice. We define evaluation as an autonomous mapping from evidence to judgments, and show that interactive evaluation changes both sides of this mapping: the evidence becomes interaction-generated trajectories, while the evaluation procedure must assess process, recoverability, coordination, robustness, and system-level performance. Building on this definition, we propose a two-axis taxonomy, derive design principles and reporting standards, examine representative scenarios, and analyze how longstanding evaluation challenges reappear at the trajectory level.
arXiv:2605.17827v1 Announce Type: new Abstract: Generative analysis often models multi-domain observations as nonlinear mixtures of domain-invariant content variables and domain-specific style variables. Identifying both factors from unpaired domains enables tasks such as domain transfer and counterfactual data generation. Prior work establishes identifiability under (block-wise) statistical independence between content and style, or via sparse Jacobian assumptions on the nonlinear mixing function, but such conditions can be restrictive in practice. In this work, we introduce content-style differential independence (CSDI), an alternative structural condition requiring that infinitesimal variations in content and style induce orthogonal directions on the data manifold, thereby enabling identifiability even when content and style are dependent and the Jacobian is dense. We operationalize this condition through a blockwise orthogonality constraint on the Jacobian subspaces associated with content and style. To support high-dimensional generative models, we design a stochastic regularizer based on numerical Jacobian approximation, enabling scalable training in settings such as high-resolution image generation. Experiments across multiple datasets corroborate the identifiability analysis and demonstrate practical benefits on counterfactual generation and domain translation.
arXiv:2605.17655v1 Announce Type: new Abstract: The European Commission recently introduced new regulation to harmonize transparency reporting of large online platforms under the Digital Services Act (DSA). Here, we present the first systematic evaluation of transparency reporting data quality after this normative change, for the eight largest social media platforms in the European Union. In detail, we run a set of large-scale quantitative analyses on key reporting dimensions, followed by a structured comparative assessment across platforms and reporting mechanisms. Among our findings is that: (i) the analyzed platforms had varying degrees of compliance and data quality, but all exhibited issues on data formatting, timeliness, consistency, and completeness; (ii) some platforms employed differing reporting procedures across mechanisms, which caused them to submit contrasting information; (iii) despite the harmonization, a number of issues still prevent interoperability between reporting mechanisms; and (iv) many of the previously identified issues with transparency reporting are still unresolved. We conclude by discussing implications for transparency auditing and proposing key targeted improvements to strengthen the reliability and interoperability of DSA transparency reporting.
arXiv:2605.17467v1 Announce Type: new Abstract: Large language model-driven multi-agent systems (LLM-MAS) excel at complex tasks, yet unreliable agents remain a key bottleneck to system-level reliability. Automatic failure attribution is therefore critical, but existing approaches, such as direct prediction of agent-error pairs and agent-first failure attribution, rely on local logs of agents and miss global failures that only manifest over full interaction trajectories, such as cross-step inconsistencies and inter-agent coordination errors. Moreover, directly predicting failures induces a large combinatorial search space, hindering fine-grained attribution. To address these challenges, we propose VerifyMAS, a hypothesis verification framework for agent failure attribution. Instead of directly predicting faulty agents and error types, VerifyMAS formulates and verifies failure hypotheses against full trajectories. This verification-based approach decomposes attribution into trajectory-level error validation and fine-grained agent localization, providing an error-first attribution approach that captures global failure patterns while substantially reducing the search space. We further introduce a hypothesis-based data construction strategy grounded in a structured error taxonomy and fine-tune a specialized LLM verifier model for trajectory-level failure verification and agent attribution. Experiments on Aegis-Bench and Who&When show that VerifyMAS consistently improves diverse backbone models, including open-source Qwen and API-based GPT models, outperforming prior methods without sacrificing inference efficiency for long multi-agent trajectories.
arXiv:2605.13745v2 Announce Type: replace Abstract: Multimode intracavity frequency-doubled lasers can reach states of amplitude noise suppression orders of magnitude beyond the predictions of independent-mode partition statistics. We show that the chi2 coupled-wave dynamics in the doubling crystal admit a Lyapunov functional whose monotone decrease under each crystal pass establishes a constant-intensity manifold as the per-pass descent target of the mode dynamics. We confirm the mechanism in an intracavity frequency-doubled Nd:YVO4-LBO laser, observing a 100 fold contrast between full and Fabry-Perot-filtered output noise at fixed detector bandwidth, well beyond the statistical-averaging baseline. The mechanism rests on the algebraic structure of the coupling, a coherent superposition of oscillators sharing a quadratic dissipative channel, and is therefore a candidate for analogous noise-suppression effects in other coupled oscillator systems with the same algebraic form.
arXiv:2605.14787v2 Announce Type: replace Abstract: Composed Image Retrieval (CIR) is a multimodal retrieval task where a query consists of a reference image and a textual modification, and the goal is to retrieve a target image satisfying both. In principle, strong performance on CIR benchmarks is assumed to require multimodal composition, i.e., combining complementary information from reference image and textual modification. In this work, we show that this assumption does not always hold. Across four widely used CIR benchmarks and eleven Generalist Multimodal Embedding models, a large fraction of queries can be solved using a single modality (from 32.2% to 83.6%), revealing pervasive unimodal shortcuts. Thus, high CIR performance can arise from unimodal signals rather than true multimodal composition. To better understand this issue, we perform a two-stage audit. First, we identify shortcut-solvable queries through cross-model analysis. Second, we conduct human validation on 4,741 shortcut-free queries, of which only 1,689 are well-formed, with common issues including ambiguous edits and mismatched targets. Re-evaluating models on this validated subset reveals qualitatively different behaviour: queries can no longer be solved with a single modality, and successful retrieval requires combining both inputs. While accuracy decreases, reliance on multimodal information increases. Overall, current CIR benchmarks conflate shortcut-solvable, noisy, and genuinely compositional queries, leading to an overestimation of model capability in multimodal composition.
arXiv:2401.14278v3 Announce Type: replace Abstract: Blockchain performance has historically faced challenges posed by the throughput limitations of consensus algorithms. Recent breakthroughs in research have successfully alleviated these constraints by introducing a modular architecture that decouples consensus from execution. The move toward independent optimization of the consensus layer has shifted attention to the execution layer. While concurrent transaction execution is a promising solution for increasing throughput, practical challenges persist. Its effectiveness varies based on the workloads, and the associated increased hardware requirements raise concerns about undesirable centralization. This increased requirement results in full nodes and stragglers synchronizing from signed checkpoints, decreasing the trustless nature of blockchain systems. In response to these challenges, this paper introduces Chiron, a system designed to extract execution hints for the acceleration of straggling and full nodes. Notably, Chiron achieves this without compromising the security of the system or introducing overhead on the critical path of consensus. Evaluation results demonstrate a notable speedup of up to 30%, effectively addressing the gap between theoretical research and practical deployment. The quantification of this speedup is achieved through realistic blockchain benchmarks derived from a comprehensive analysis of Ethereum and Solana workloads, constituting an independent contribution.
arXiv:2605.18697v1 Announce Type: new Abstract: Compound AI applications, which compose calls to ML models using a general-purpose programming language like Python, are widely used for a variety of user-facing tasks, from software engineering to enterprise automation, making their end-to-end latency a critical bottleneck. In contrast to traditional applications, execution time is dominated by the external components, which cannot be handled by traditional language optimization systems, like optimizing compilers. To address this problem, we develop PopPy, a system that can uncover parallelization opportunities in Python applications that invoke these heavy external components, including those used in compound AI applications. PopPy supports a very expressive fragment of Python and requires minimal developer input to uncover parallelism. It combines an ahead-of-time compiler with a runtime, addressing three key challenges in extracting parallelism from Python applications: language complexity, dynamic dispatch, and variable mutation. On a set of real-world compound AI applications, PopPy achieves up to $6.4\times$ speedups in end-to-end execution time compared to standard Python execution while preserving the sequential program semantics.
arXiv:2605.12451v2 Announce Type: replace Abstract: Continual Panoptic Segmentation (CPS) requires methods that can quickly adapt to new categories over time. The nature of this dense prediction task means that training images may contain a mix of labeled and unlabeled objects. As nothing is known about these unlabeled objects a priori, existing methods often simply group any unlabeled pixel into a single "background" class during training. In effect, during training, they repeatedly tell the model that all the different background categories are the same (even when they aren't). This makes learning to identify different background categories as they are added challenging since these new categories may require using information the model was previously told was unimportant and ignored. Thus, we propose a Future-Targeted Contrastive and Repulsive (FuTCR) framework that addresses this limitation by restructuring representations before new classes are introduced. FuTCR first discovers confident future-like regions by grouping model-predicted masks whose pixels are consistently classified as background but exhibit non-background logits. Next, FuTCR applies pixel-to-region contrast to build coherent prototypes from these unlabeled regions, while simultaneously repelling background features away from known-class prototypes to explicitly reserve representational space for future categories. Experiments across six CPS settings and a range of dataset sizes show FuTCR improves relative new-class panoptic quality over the state-of-the-art by up to 28%, while preserving or improving base-class performance with gains up to 4%.