arXiv:2607.06504v1 Announce Type: new Abstract: Recent years have witnessed the emergence of multivariate modeling using time series foundation models (TSFMs), which achieve advanced zero-shot generalization. Modern multivariate TSFMs are predominantly pretrained on multivariate synthetic data, which is easier to scale but may fail to capture the complex temporal dynamics and cross-variable relationships present in real-world time series. This raises a key question: Whether and to what extent the leading TSFMs trained with the real-world corpus perform better than those trained with synthetic data? To answer this, we establish the RMISC corpus, a considerably large-scale, high-quality, openly accessible, real-world, and multivariate time series archive that contains around 200 datasets and 142 billion time points across diverse domains. Furthermore, we pretrain four advanced TSFMs on univariate, synthetic multivariate, and real-world multivariate data and evaluate their zero-shot generalization capabilities on standard in-distribution and out-of-distribution benchmarks. Experimental results show that incorporating real-world multivariate data predominantly improves the generalization performance for both univariate and multivariate TSFMs. These results provide a deeper understanding of how real-world multivariate data contributes to the development of stronger TSFMs.
Science Journals
arXiv:2503.10954v3 Announce Type: replace Abstract: Large Language Models (LLM) can solve *any* computational problem *without* an algorithm in a runtime *independent* of the computational complexity of that problem. Instead of specifying precisely how to solve problem instance using *programming*, we ask an LLM to solve the problem instance using *prompting*. Outputs are sampled from a distribution rather than generated procedurally. In this vision paper, we explore the challenges and opportunities of this new form of computation and observe that its capabilities and limits *cannot* be understood within the classic, rationalist framework of computation. Hence, we appeal to the software engineering (SE) community to develop the foundations and techniques required to analyze the properties of this "empirical computation" as it generates solutions to computational problems: How can we analyze and improve the correctness of LLMs solving a computational problem in the general, in the problem-specific, or in the instance-specific? What are the properties and fundamental limits of empirical computation? This paper aims to establish empirical computation as a field in SE that is timely and rich with interesting problems.
arXiv:2601.20753v4 Announce Type: replace Abstract: Preference-Conditioned Policy Learning (PCPL) in Multi-Objective Reinforcement Learning (MORL) approximates diverse Pareto-optimal solutions by conditioning a single policy on user-specified preferences, enabling run-time adaptation to arbitrary trade-offs without retraining. However, existing PCPL benchmarks are largely restricted to toy tasks and fixed environments, limiting their realism and scalability. To address this gap, we introduce GraphAllocBench, a flexible benchmark built on CityPlannerEnv, a novel graph-based resource allocation sandbox inspired by city management. GraphAllocBench provides a rich suite of problems with customizable objective functions, varying preference conditions, complex Pareto Fronts, and high-dimensional scalability. We further propose two supplementary metrics -- Proportion of Non-Dominated Solutions (PNDS) and Ordering Score (OS) -- that capture prediction reliability and preference consistency while complementing the widely used hypervolume metric. Through experiments with several state-of-the-art PCPL algorithms and our own MLP and graph-aware PCPL-PPO baseline, we show that GraphAllocBench exposes distinct failure modes that hypervolume alone does not capture but our supplementary metrics reveal, while motivating graph-based approaches such as Graph Neural Networks (GNNs) for scaling to complex, high-dimensional allocation tasks. By letting users freely vary objectives, preferences, and allocation rules, GraphAllocBench serves as a versatile and extensible testbed for advancing PCPL.
arXiv:2601.21135v2 Announce Type: replace Abstract: Temporal causal representation learning methods assume that causal mechanisms switch instantaneously between discrete domains, yet real-world systems often exhibit continuous mechanism transitions. For example, a vehicle's dynamics evolve gradually through a turning maneuver, and human gait shifts smoothly from walking to running. We formalize this setting by modeling transitional mechanisms as convex combinations of finitely many atomic mechanisms, governed by time-varying mixing coefficients. Our theoretical contributions establish that both the latent causal variables and the continuous mixing trajectory are jointly identifiable. We further propose TRACE, a Mixture-of-Experts framework where each expert learns one atomic mechanism during training, enabling recovery of mechanism trajectories at test time. This formulation generalizes to intermediate mechanism states never observed during training. Experiments on synthetic and real-world data demonstrate that TRACE recovers mixing trajectories with up to 0.99 correlation, substantially outperforming discrete-switching baselines.
arXiv:2607.05628v1 Announce Type: new Abstract: Accurate and efficient classification of thoracic diseases in chest X-ray (CXR) images is crucial for timely diagnosis and treatment. However, the presence of multiple pathologies with overlapping visual characteristics poses significant challenges for automated classification systems. In this study, we propose two novel hierarchical multi-label classification techniques, namely the loss-based and logit-based methods, to address these challenges by leveraging the hierarchical relationships among different thoracic pathologies. The loss-based technique integrates hierarchical information directly into the optimization process, while the logit-based method adjusts the predicted probabilities of each class based on its parent class in the disease taxonomy. We evaluate the performance of both techniques using three large-scale CXR datasets: CheXpert (224,316 CXRs), PADCHEST (160,000 CXRs), and NIH (112,120 CXRs). The experimental results demonstrate significant improvements in accuracy, AUC, and F1 scores compared to the baseline method across various pathologies. The logit-based and loss-based methods improve accuracy by 12\% and 11\%, AUC by 13\% and 10\%, and F1 scores by 24\% and 12\%, respectively compared to the baseline. These results represent a substantial improvement over the baseline method. Furthermore, we conduct a comprehensive statistical analysis to validate the robustness and reliability of the proposed techniques. The integration of domain-specific hierarchical knowledge not only enhances the classification performance but also provides a more interpretable output for clinical decision support. Our findings highlight the potential of hierarchical multi-label classification in advancing computer-aided diagnosis systems for chest radiography.
arXiv:2607.05632v1 Announce Type: new Abstract: The automotive industry's shift toward software-driven systems has increased system complexity and raised the importance of effective requirement intake and refinement for correctness, compliance, development speed, and systematic reuse. Although prior research has proposed techniques for improving requirement quality, limited empirical evidence exists on how stakeholder-level requirements are evaluated, refined, and transformed into product-level requirements in industrial automotive practice. This paper presents a large-scale empirical study based on an industrial dataset from Infineon, comprising 8,082 stakeholder requirements and 5,870 product requirements enriched with traceability links, decision outcomes, deviation rationales, and domain references. Using a mixed-methods approach, we combine quantitative analyses of requirement structures, decision distributions, and mapping patterns with qualitative analyses of rationales, referenced specifications, and software- and hardware-related artifacts. We investigate structural and contextual differences between stakeholder and product requirements, factors influencing acceptance, rejection, and approval with deviation, and the nature of stakeholder-to-product refinement. The results reveal systematic differences across abstraction levels and show that refinement complexity is driven primarily by architectural scope and missing contextual information rather than linguistic verbosity. We further derive a taxonomy of stakeholder-product mapping patterns and relate these patterns to differing refinement effort. The findings provide concrete insight into industrial requirements intake and refinement practices and identify actionable opportunities for improving intake validation, deviation management, and tool-supported contextual enrichment to support faster and more reusable automotive product development.
arXiv:2607.05633v1 Announce Type: new Abstract: In this work we design a multi-layer model of composite contact lens drug release. Such lenses have been designed by encapsulating drug-polymer films in contact lens hydrogels. Composite lenses can promote sustained discharge of drug and achieve near zero-order release kinetics, thus surpassing other ocular delivery methods that are limited by short residence times and an undesirable initial burst release. Our model is informed by in vivo data, and includes three coupled partial differential equation layers to simulate the composite lens. We mathematically investigate the effect of composite contact lens design characteristics on the time to $50\%$ therapeutic drug release ($t_{50}$) in the vial, eye, and blister pack settings. In the eye setting, we incorporate our prior model that considers the effect of many blinks on the pre- and post-lens tear film drug concentrations. We simulate drug cumulative release profiles and study the variability of $t_{50}$ across: (1) the ratio of the drug-polymer film to hydrogel diffusion coefficients, (2) the centerline of the polymer film within the hydrogel, and (3) the polymer film thickness. In the blister pack setting, we study storage questions that may inform future commercial design. This work may help medical professionals better understand the mechanics of contact lens drug delivery and predict targeted tissue transport of ophthalmic drugs.
arXiv:2512.07997v2 Announce Type: replace Abstract: Gestures are an integral part of our daily interactions with the environment. Hand gesture recognition (HGR) is the process of interpreting human intent through various input modalities, such as visual data (images and videos) and bio-signals. Bio-signals are widely used in HGR due to their ability to be captured non-invasively via sensors placed on the arm. Among these, surface electromyography (sEMG), which measures the electrical activity of muscles, is the most extensively studied modality. However, less-explored alternatives such as inertial measurement units (IMUs) can provide complementary information on subtle muscle movements, which makes them valuable for gesture recognition. In this study, we investigate the potential of using IMU signals from different muscle groups to capture user intent. Our results demonstrate that IMU signals contain sufficient information to serve as the sole input sensor for static gesture recognition. Moreover, we compare different muscle groups and check the quality of pattern recognition on individual muscle groups. We further found that tendon-induced micro-movement captured by IMUs is a major contributor to static gesture recognition. We believe that leveraging muscle micro-movement information can enhance the usability of prosthetic arms for amputees. This approach also offers new possibilities for hand gesture recognition in fields such as robotics, teleoperation, sign language interpretation, and beyond.
arXiv:2512.11001v2 Announce Type: replace Abstract: The proliferation of large language models (LLMs) has accelerated the adoption of agent-based data pipelines. Yet current approaches remain ad hoc, relying on fixed structures, predefined LLMs, and single execution engines, without systematic optimization across heterogeneous data sources and engines. This paper presents NOMA, a query optimization framework for multi-agent data pipelines. We argue that optimizing agentic pipelines is a fundamentally different query optimization problem, with central challenges: (i)~a multi-dimensional search and objective space, where topology, model, and engine choices must be optimized jointly across latency, cost, and accuracy; (ii)~a variable pipeline topology; (iii)~the co-existence of diverse data models, leaving no common operator algebra; and (iv)~the significant cost of executing these pipelines. Our controlled experiment over a real-world 10-agent pipeline reveals extreme variance (153x cost, 5x latency, 25% quality) and that optimal plans are heterogeneous configurations no user would construct manually. Our analysis of real deployments confirms these inefficiencies are systematic. We present \system{} as an integrated optimization loop in which plan generation, cost estimation, runtime refinement, and semantic caching reinforce one another across executions, setting a community-wide research agenda on query optimization for multi-agent systems.
arXiv:2607.06424v1 Announce Type: new Abstract: X-ray fluorescence (XRF) spectroscopy is a key modality for material analysis in cultural heritage. However, automated learning from XRF spectra remains challenging: XRF spectra are complex one-dimensional signals composed of sharp elemental peaks, broader structures, and background variations that are not taken into account by existing learning-based models. This paper introduces XRFormer, a transformer architecture tailored to XRF spectra through a multiscale convolutional tokenizer that injects locality and multi-resolution inductive biases before global self-attention. The tokenizer progressively reduces spectral resolution while increasing embedding dimensionality, and the resulting token sequence is processed by a standard transformer encoder. We further investigate self-supervised pretraining for XRF representation learning using Masked Spectral modeling (MSM) and a physics-informed Peak Presence Prediction (PPP) objective. Experiments on the Pigments Checker STANDARD v.5 dataset for pigment identification and unmixing show that XRFormer consistently outperforms ViT, SpectralFormer (with and without CAF), and a 1D-CNN baseline for pigment identification. For pigment unmixing, XRFormer achieves robust abundance estimation while maintaining significantly higher parameter efficiency than SpectralFormer, operating at a lower token resolution (128 vs. 512 tokens) and with less than half the number of parameters (1.5M vs. 3.37M). MSM yields consistent gains across both tasks, while PPP further enhances performance for both identification and unmixing when tuned with an appropriate peak prominence. These results highlight multiscale, modality-aware tokenization as an effective and parameter efficient foundation for transformer-based XRF modeling under data-limited conditions. A GitHub repository is provided at https://github.com/sofiane1010/XRFormer.
arXiv:2607.05658v1 Announce Type: new Abstract: Global Navigation Satellite Systems (GNSS), best known for positioning, also serve weather science, as atmospheric water vapour delays their signals. This delay, the Zenith Wet Delay (ZWD), is a direct, all-weather measure of column moisture. Although assimilated into numerical weather prediction for decades, ZWD is not yet used by leading machine learning weather models (MLWM), despite addressing a known deficiency: the underestimation of severe precipitation. Here we present the first integration of GNSS-derived ZWD into Aurora, a state-of-the-art weather foundation model. Our extended Aurora learns ZWD with skill comparable to its pretrained variables. More importantly, including ZWD systematically improves forecasts when fine-tuning for six-hour accumulated precipitation. Gains grow with severity, reaching an 8.8\% increase in Equitable Threat Score at the 99th percentile, while the precipitation power spectrum becomes more realistic at synoptic and planetary scales. Direct GNSS observations therefore encode information that MLWM can exploit for high-impact precipitation.
arXiv:2504.16595v2 Announce Type: replace Abstract: Packing objects efficiently is a fundamental problem in logistics, warehouse automation, and robotics. When dealing with highly diverse 3D objects (household or grocery items), closed-form solutions are infeasible, and heuristic or model-free Reinforcement Learning~(RL) methods tend to focus solely on geometric optimization, relying on exhaustive searches of the discretized solution space. This leads to long training times (for pure RL) and high latency (heuristics), limited transferability to robotic scenarios, and ultimately ignores object characteristics (fragility, deformability) and human preferences. We propose HERB, a human-augmented RL framework for packing irregular objects, the first to explore the potential of learning from human demonstrations to solve this complex task. It leverages human demonstrations of packing strategies, which inherently exhibit latent factors such as space optimization, stability, and object properties that are difficult to model explicitly. The human-expert data is combined with RL exploration to provide the placement of each object inside the container. Experimental results show that our method outperforms heuristic, purely RL-based, and imitation learning approaches in packing efficiency and latency. Qualitative results highlight that our packing strategy produces more stable, human-like arrangements, which we expect to be more appropriate and widely accepted. Finally, we demonstrate the real-world feasibility of our method on a robotic system.
arXiv:2601.21803v2 Announce Type: replace Abstract: Retrieval-Augmented Generation (RAG) systems combine dense retrievers and language models to ground their outputs in external documents. However, the interaction between these components remains opaque, creating challenges for deployment in high-stakes domains. We present RAG-E, an end-to-end explainability framework that quantifies retriever-generator alignment through mathematically grounded attribution methods. Our approach adapts Integrated Gradients for retriever analysis, proposes a Monte Carlo-stabilized Shapley Value approximation for generator attribution, and introduces the Weighted Attribution-Relevance Gap (WARG) metric to measure how closely the generator's document usage aligns with retriever rankings. Experiments on PopQA, QAMPARI, and TREC CAST datasets reveal substantial misalignment: depending on the model and setting, generators often ignore top-ranked documents and rely on documents ranked as less relevant. We show that WARG captures retriever-generator alignment better than Pearson and Spearman correlations and can serve as an indicator of RAG performance. RAG-E and WARG provide a practical framework for auditing this interaction, enabling more reliable and transparent RAG systems.
arXiv:2512.13919v3 Announce Type: replace Abstract: This work shows how adaptivity can enhance value realization of digital twins in civil engineering. We focus on adapting the state transition models within digital twins represented through probabilistic graphical models. The bi-directional interaction between the physical and virtual domains is modeled using dynamic Bayesian networks. By treating state transition probabilities as random variables endowed with conjugate priors, we enable hierarchical online learning of transition dynamics from a state to another through effortless Bayesian updates. We provide the mathematical framework to account for a larger class of distributions with respect to the current literature on digital twins. To compute dynamic policies with precision updates we solve parametric Markov decision processes through reinforcement learning. The proposed adaptive digital twin framework enjoys enhanced personalization, increased robustness, and improved cost-effectiveness. We assess our approach on a case study involving structural health monitoring and maintenance planning of a railway bridge.
arXiv:2605.30794v2 Announce Type: replace Abstract: Multimodal Large Language Models (MLLMs) have demonstrated significant achievements in general visual question answering (VQA) tasks. However, they remain brittle on mechanical engineering drawings, where high annotation density and weak domain knowledge, compounded by unreliable spatial relation reasoning under strict projection rules and geometric constraints, make decisive cues easy to miss and frequently lead to wrong answers. To bridge this gap, we introduce the first comprehensive mechanical drawing understanding dataset, MechVQA, created through a semi-automated construction and quality-control pipeline. MechVQA contains 3.3k high-density pictures with 21K question-answer pairs, spanning 10 different fine-grained tasks across three capability levels: Recognition, Reasoning, and Judging, providing a testbed to evaluate and improve MLLM understanding on real-world mechanical drawings. On top of MechVQA, we then develop the MechVL model through a multi-stage training paradigm, building a strong domain-specialized baseline. Extensive experimental results demonstrate that MechVL outperforms the strongest closed-source baseline by 7.57 percentage points on the MechVQA total score, significantly enhancing mechanical drawing understanding ability and providing a reusable foundation for deploying MLLMs in mechanical design and inspection scenarios.
arXiv:2605.30796v5 Announce Type: replace Abstract: This paper presents a rigorous convergence analysis for the lightning plus polynomial approximation scheme, which employs rational approximations constructed with preassigned tapered, exponentially clustered poles. This pole placement strategy was originally introduced by Trefethen and his collaborators for the resolution of corner singularities. Ample numerical results indicate that this scheme achieves root-exponential convergence, and in particular attains the same optimal convergence rate as the best rational approximation to $x^\alpha$ on $[0,1]$ established by Stahl.% which is conjectured in [SIAM J. Numer. Anal., 61:2580-2600, 2023]. In this work, we establish optimal root-exponential convergence for the class of prototype functions of the form $g(z)z^\alpha$ or $g(z)z^\alpha\log z$, where $g$ is analytic on a neighborhood of the sector domain. These results confirm the validity of Conjectures 3.1 and 5.3 stated in [SIAM J. Numer. Anal., 61:2580-2600, 2023], and demonstrate that the choice $\sigma_{\mathrm{opt}} =\frac{\sqrt{2(2 - \beta)}\pi}{\sqrt{\alpha}}$ achieves the theoretically optimal convergence rate $\mathcal{O}\left(e^{-\sqrt{2(2 - \beta)N\alpha}\pi}\right)$. Notably, for the specific case of $\beta = 0$, the scheme recovers Stahl's optimal convergence rate for $x^\alpha$. Furthermore, working within the decomposition framework for corner domains proposed by Gopal and Trefethen, this paper provides a rigorous proof of optimal root-exponential convergence for lightning plus polynomial approximation problems on corner domains, and explicitly derives the optimal pole clustering parameter.
arXiv:2603.06924v2 Announce Type: replace Abstract: In classical Informative Path Planning (C-IPP), robots are typically modeled as mobile sensors that acquire digital measurements such as images or radiation levels. In this model, since making a measurement leaves the robot's physical state unchanged, the cost of traversing an edge remains static regardless of when it is traversed. This is a natural assumption for many missions, but does not extend to settings involving physical sample collection, where each collected sample adds mass and increases the energy cost of all subsequent motion. As a result, IPP formulations that ignore this coupling between information gain and load-dependent traversal cost can produce plans that are distance-efficient but energy-suboptimal, collecting fewer samples and less data than the energy budget would permit. In this paper, we first introduce Load-aware Informative Path Planning (LIPP), a strict generalization of C-IPP that explicitly models this coupling, with C-IPP recovered as the special case of zero sample mass. We then formulate LIPP as a Mixed-Integer Quadratic Program (MIQP) that jointly optimizes visitation location, order, and per-location sampling count under an energy budget. We further derive theoretical bounds on the path-length increase of LIPP relative to C-IPP, characterizing the trade-off for improved energy efficiency. Finally, through extensive simulations across 2,000 diverse mission scenarios, we demonstrate that LIPP progressively achieves higher uncertainty reduction per unit energy as sample mass increases.
arXiv:2512.17143v3 Announce Type: replace Abstract: Photographs of people taken by professional photographers typically present the person in beautiful lighting, with an interesting pose, and flattering quality. This is unlike common photos people take of themselves in uncontrolled conditions. In this paper, we explore how to canonicalize a person's 'in-the-wild' photograph into a controllable, high-fidelity avatar -- reposed in a simple environment with standardized minimal clothing. A key challenge is preserving the person's unique whole-body identity, facial features, and body shape while stripping away the complex occlusions of their original garments. While a large paired dataset of the same person in varied clothing and poses would simplify this, such data does not exist. To that end, we propose two key insights: 1) Our method transforms the input photo into a canonical full-body UV space, which we couple with a novel reposing methodology to model occlusions and synthesize novel views. Operating in UV space allows us to decouple pose from appearance and leverage massive unpaired datasets. 2) We personalize the output photo via multi-image finetuning to ensure robust identity preservation under extreme pose changes. Our approach yields high-quality, reposed portraits that achieve strong quantitative performance on real-world imagery, providing an ideal, clean biometric canvas that significantly improves the fidelity of downstream applications like Virtual Try-On (VTO).
SecureCode: A Production-Grade Multi-Turn Dataset for Training Security-Aware Code Generation Models
arXiv:2512.18542v3 Announce Type: replace Abstract: AI coding assistants produce vulnerable code in 45\% of security-relevant scenarios~\cite{veracode2025}, yet no public training dataset teaches both traditional web security and AI/ML-specific defenses in a format suitable for instruction tuning. We present SecureCode, a production-grade dataset of 2,185 multi-turn security training examples spanning two domains: web application security (1,435 examples covering the OWASP Top 10 2021 across 11 languages and 9 frameworks, 100\% grounded in documented CVEs and security incidents) and AI/ML security (750 examples covering all 10 OWASP LLM Top 10 2025 categories across more than 40 frameworks, including LangChain, OpenAI, and Hugging Face). Every example follows a 4-turn conversational structure -- feature request; vulnerable and secure implementations with attack demonstrations; advanced probing; and defense-in-depth operational guidance -- designed for direct use in instruction tuning pipelines. Quality assurance combines automated structural validation with multi-agent review from seven specialist AI perspectives (more than 10{,}500 assessments) and an 8-phase remediation pipeline, producing a rubric-calibrated mean quality score of 93.8/100 ($\sigma = 0.93$) for the AI/ML component. Each example provides SIEM integration strategies, infrastructure hardening recommendations, and testing approaches using production frameworks. We release the unified dataset on Hugging Face with domain-specific loading configurations (web, aiml, default), alongside eight fine-tuned open-source models (3B--20B parameters, QLoRA), and an evaluation framework with four security-specific metrics. To our knowledge, SecureCode is the first public dataset that jointly provides OWASP Top 10 2021 web coverage and OWASP LLM Top 10 2025 AI/ML coverage in a unified conversational schema suitable for instruction tuning.
arXiv:2602.22600v2 Announce Type: replace Abstract: Training selects for behavior, not circuitry: many weight configurations can implement the same function. Studying any single trained neural network thus risks describing accidents of one training run rather than the computation itself. This work shifts focus from what transformers happen to do to what they must do by extracting algorithmic cores, compact subspaces that are necessary and sufficient for a task and that recur across independently trained models. Here, Algorithmic Core Extraction (ACE) is introduced to isolate these subspaces, causally validate them, and recover the algorithms they implement across settings ranging from synthetic tasks to large-scale pretrained models. Markov-chain transformers embed three-dimensional cores in nearly orthogonal subspaces yet recover identical transition spectra. Modular-addition transformers form compact cyclic cores at grokking that later inflate under continued regularization, redundantly distributing the same computation across many functionally equivalent modes. This functional redundancy is found to accelerate the transition from memorization to generalization, yielding an inverse scaling law for grokking time. In six language models spanning more than two orders of magnitude in scale (GPT-2 Small/Medium/Large, LLaMA-3.1, Gemma-2, and Qwen2.5), subject-verb agreement is governed by a single, steerable axis that aligns across architectures. Flipping this axis inverts grammatical number throughout open-ended generation. Together these results suggest that beneath the apparent complexity of trained transformers lies a simpler, shared computational structure, and that targeting invariants rather than parameterizations may offer a more tractable path to mechanistic understanding and control. Code: https://github.com/joshseth/cores
arXiv:2603.15600v2 Announce Type: replace Abstract: Accurate process supervision remains a critical challenge for long-horizon robotic manipulation. A primary bottleneck is that current video MLLMs, trained primarily under a Supervised Fine-Tuning (SFT) paradigm, function as passive "Observers" that recognize ongoing events rather than evaluating the current state relative to the final task goal. In this paper, we introduce PRIMO R1 (Process Reasoning Induced Monitoring), a 7B framework that transforms video MLLMs into active "Critics". We leverage outcome-based Reinforcement Learning to incentivize explicit Chain-of-Thought generation for progress estimation. Furthermore, our architecture constructs a structured temporal input by explicitly anchoring the video sequence between initial and current state images. Supported by the proposed PRIMO Dataset and Benchmark, extensive experiments across diverse in-domain environments and out-of-domain real-world humanoid scenarios demonstrate that PRIMO R1 achieves state-of-the-art performance. Quantitatively, our 7B model achieves a 50% reduction in the mean absolute error of specialized reasoning baselines, demonstrating significant relative accuracy improvements over 72B-scale general MLLMs. Furthermore, PRIMO R1 exhibits strong zero-shot generalization on difficult failure detection tasks. We establish state-of-the-art performance on RoboFail benchmark with 67.0% accuracy, surpassing closed-source models like OpenAI o1 by 6.0%.
arXiv:2605.31480v2 Announce Type: replace Abstract: Do neural models, such as Large Language Models, genuinely acquire compositional abilities for interpretation of natural language? When we talk about semantic interpretation, we can distinguish two complementary aspects: establishing what an expression refers to in the world (which we call the Extensional task) and representing its sense in a structured way (which we call the Intensional task). We evaluate LLMs and humans on both tasks in the setting of the Personal Relation Task (Paperno 2022) in which, given a universe of people and their relationships with each other, one is asked to interpret a noun phrase such as "Amber's parent's friend". Here, for the Intensional task, the answer is the formula "friend(parent(amber))", and for the Extensional task, the person. We find that humans and LLMs show opposite strengths: humans perform better on Extensional than Intensional tasks, and LLMs vice versa. Our methodology brings greater nuance to the understanding of compositional abilities in modern machine learning models. Our results support the notion that the lack of referential grounding in LLM training is a crucial missing component in mimicking human-like language understanding.
arXiv:2606.00476v2 Announce Type: replace Abstract: Do LLM agents act on the reasoning they state? This question of process fidelity is central to LLM-based social simulation, yet hard to measure where no reference for correct behavior exists. We study it in a controlled setting: a Texas Poker simulator with a verifiable reference action for every decision by splitting the faithfulness gap into two steps: reasoning-to-conclusion (does the stated decision follow from the agent's own reasoning?) and conclusion-to-action (does the agent execute what it states?). The two steps behave very differently. Conclusion-to-action is reliable: inconsistency is 0.7% for Claude Haiku 4.5 and 1.4% for DeepSeek-Reasoner once the conclusion is read from an explicit tag, whereas free-text conclusion extraction reports 22-26%. Reasoning-to-conclusion is where fidelity frays, but not through a single dominant failure. In a step-level diagnostic the agent's errors split roughly evenly between bad inputs, borderline cases, and rule misapplication deriving a conclusion that contradicts the agent's own restated rule from inputs it estimated correctly. This composition is model-dependent: rule misapplication accounts for a third of Haiku's interpretable errors but only 8% of DeepSeek's. The one robust signal is directional: when an agent does misapply its own stated rule, it almost always (99.5% for Haiku) errs in the risk-averse direction. The override is partly hedging behavior, not a capability limit: instructing the agent to apply the rule mechanically halves the misapplication rate (13.9% to 6.8% of decisions) and raises adherence by eight points. Process-fidelity evaluation should therefore elicit machine-checkable conclusions and probe for directional biases rather than assume a single upstream failure mode, lest it conflate measurement noise with model behavior.
arXiv:2606.00550v2 Announce Type: replace Abstract: The persistent challenge in scaling authentic manipulator education within university laboratories is a structural dichotomy: commercial digital twins are often cost-prohibitive and rigidly scripted, whereas open-source robotics middleware (ROS) imposes steep technical and syntax barriers for novices. To resolve this logistical and educational friction, this paper proposes a scalable four-tier communication architecture tailored for sustainable robotic curricula. Rather than focusing on software application design, our study examines the underlying data exchange mechanisms required to bridge visual conceptual environments with physical robotic endpoints, utilizing the Graphical Open-Source Platform (GOSP) as a reference implementation. Our work details the framework's technical integration of 3D visual armature modeling with a robust ROS middleware backend, emphasizing the serialization, routing, and encapsulation of intricate communication routines. Preliminary sim-to-real validation using multi-axis spatial trajectories confirms that encapsulating these communication pipelines provides sufficient fidelity and a hardware-agnostic pathway. By bridging virtual design and physical execution, our architectural blueprint offers a viable infrastructure for engineering education.
arXiv:2606.00552v2 Announce Type: replace Abstract: Multi-robot systems (MRS) increasingly offload compute-intensive perception tasks to edge nodes to meet strict time-sensitive Quality-of-Service (QoS) constraints. However, static task orchestration on a shared edge node can severely degrade QoS due to network latency, jitter, and edge-resource contention. We present a pilot edge-centric MRS testbed using Raspberry Pi nodes to evaluate a camera-to-manipulator pipeline under three modes: local execution, static offloading, and a QoS-aware Adaptive Task Placement (ATP) controller. ATP scores candidate placements using a multi-metric cost (normalized latency, CPU utilization, and switching overhead) over two-second control windows. The closed-loop visual servoing testbed is instrumented with sub-millisecond clock synchronization, network emulation, and detailed monitoring of multiple metrics across nodes to capture realistic jitter. Experimental results under compute-stress and network-fault scenarios show that static edge offloading reduces on-board CPU load but amplifies tail latency and deadline misses. In contrast, the QoS-aware ATP controller, by switching task placement based on measured latency and utilization thresholds, consistently lowers deadline violations and tail latency. Overall, the results position ATP as a practical edge-side control primitive for MRS and concrete design guidelines for Cloud-Edge Robotics deployments within the broader cloud-fog automation, while motivating QoS-aware multi-objective workload orchestration for industrial cyber-physical systems.