arXiv:2606.17477v1 Announce Type: new Abstract: Out-of-distribution (OOD) detection in dynamic open-world environments requires a model to continually adapt to evolving data distributions while generalizing to covariate-shifted inputs and rejecting semantic-shifted OOD examples. Most existing OOD detection methods optimize only the current-step objective and do not explicitly account for how post-deployment environment changes affect future OOD behavior. In this paper, we establish a theoretical grounding for dynamic OOD detection using a reinforcement learning (RL)-guided optimizer that explicitly favors updates that reduce the semantic OOD false positive rate over time. We develop a novel augmented optimizer that uses an RL-guided correction term on top of standard gradient descent (GD) and show its improvement over both future-domain generalization and semantic-OOD rejection. We analyze temporal error decomposition in terms of model-change and environment-change generalization errors and develop a new theoretical framework for comparing the generalization errors under both GD and RL-guided optimizers.
Science Journals
arXiv:2606.17478v1 Announce Type: new Abstract: As LLMs acquire stronger reasoning capabilities, deceptive behavior becomes an increasingly serious safety concern. Existing deception monitors either score visible transcripts or derive scalar probe scores from representation vectors, leaving little inspectable evidence about why a response is suspicious. We introduce STATEWITNESS, an activation explainer for deception auditing. A separate decoder reads a target model's hidden states, then answers natural-language queries or emits structured reports about them. We evaluate STATEWITNESS on two target reasoning LLMs across seven deception datasets. STATEWITNESS reaches 0.916 mean AUROC, a relative gain of 11.6% over the best black-box text monitor and 25.0% over the best activation-probe baseline under the same evaluation protocol. When combined with existing monitors, STATEWITNESS reduces missed deceptive examples in simple threshold ensembles. Beyond scalar detection, the decoder returns query-level answers, schema reports, and token- or sentence-level evidence traces for human inspection. We view this interface as a potential building block for broader interpretability and alignment tools.
arXiv:2606.17816v1 Announce Type: new Abstract: Understanding gradient descent dynamics is key to explaining the success of over-parameterized models, where implicit bias manifests through conservation laws in gradient flow. While such laws are well understood for linear and ReLU networks, they remain largely unexplored for modern architectures. This work develops a unified framework to characterize conservation laws for contemporary models, including feedforward networks with GELU, SiLU, and SwiGLU activations, multihead attention with sinusoidal and rotary positional encodings, and Mixture-of-Experts architectures under diverse gating designs. Our theoretical findings are supported by experiments that validate the predicted invariants.
arXiv:2606.17925v1 Announce Type: new Abstract: This work extends recent developments in studying human--machine interaction by scaling from individual game-theoretic models to a societal-level model. We adopt a Graphon Mean-Field Game (GMFG) that models the interaction among four groups of internally-homogeneous but externally-heterogeneous agents in a shared environment. Our results show that parasitism can masquerade as productive learning, with knowledge distribution and actions appearing healthy while being driven by machine coupling rather than independent investigation. To detect this, we measure the direction of information flow and belief entropy of the environment, revealing that human to machine channel dominates across all scenarios, with the asymmetry intensifying under parasitism. We further demonstrate that the system exhibits coexisting mutualistic and parasitic equilibria, where environmental noise can induce a tipping point that shifts agents past the cognitive cost barrier. These emergent phenomena are not designed into any individual agent but arise from the collective interaction structure, underscoring the need to study the sociology of humans and machines holistically as a complex system.
arXiv:2606.17934v1 Announce Type: new Abstract: For PDEs naturally posed in the de Rham complex, structure-preserving mixed and saddle-point finite element discretizations typically produce indefinite linear systems. We propose a multigrid preconditioning framework that combines mass-lumped (explicitly invertible) FEEC mass matrices with transforming smoothers that map the operator to a block form with positive definite diagonal blocks, enabling Gauss-Seidel-type relaxation on the transformed system. Under mild h-uniform norm-equivalence assumptions (and for trivial topology), we prove stability of the mass-lumped systems, and by extension spectral equivalence between the mass-lumped and original FEEC operators, which motivates using multigrid cycles designed for the mass-lumped operators as preconditioners for the consistent FEEC systems. While our primary focus is on algorithmic design rather than formal convergence theory, extensive numerical experiments on the Hodge-Dirac operator, mixed Hodge-Laplacians, and a magnetostatics saddle-point system in 2D and 3D demonstrate the robustness of the approach.
arXiv:2603.10733v4 Announce Type: replace Abstract: Smooth words over an alphabet of non-negative integers $\{a,b\}$ are infinite words that are infinitely derivable, the emblematic example being the Oldenburger-Kolakoski word over $\{1,2\}$. The main way to study their language is to consider a finite version of smooth words that we call f-smooth words. In this paper we prove that the f-smooth words are exactly the factors of smooth words, and we make progress towards the conjecture of Sing that the complexity of f-smooth words over $\{a,b\}$ grows like $\Theta\left(n^{\log(a+b)/\log((a+b)/2)}\right)$: we prove it over even alphabets, we prove the lower bound over any binary alphabet and we improve the known upper bound over odd alphabets.
arXiv:2606.17966v1 Announce Type: new Abstract: Mamba-based state space models offer linear-time long-range modeling for high-resolution dense prediction, but sequential state-space propagation can attenuate boundary-sensitive and detail-sensitive responses that are critical in multi-class semantic segmentation. We propose Reload-Mamba, a semantic segmentation framework that addresses this propagation-induced response dilution through three segmentation-specific designs: (i) a boundary-supervised local detail prior that is explicitly trained with ground-truth boundary masks to identify regions requiring response restoration; (ii) a class-uncertainty-aware Reload Gate that incorporates per-pixel class entropy from a pre-reload auxiliary head as an additional gating signal, a formulation that is informative only under multi-class dense prediction; and (iii) a hierarchical multi-level Reload mechanism that applies anti-dilution refinement at three decoder levels and fuses the restored representations top-down. Built upon a ConvNeXt-Tiny encoder with a multi-scale decoder and four-directional Mamba scanning with pixel-wise directional attention, Reload-Mamba achieves 47.9% single-scale (48.9% multi-scale) mIoU on ADE20K and 83.2% single-scale mIoU on Cityscapes. With ResNet-101 + COCO pre-training under the standard DeepLab-style protocol, Reload-Mamba reaches 87.8% mIoU on PASCAL VOC 2012 val. Controlled ablations show that each of the three segmentation-specific designs contributes beyond a direct port of the prior anti-dilution architecture proposed for binarization, cumulatively improving over the direct-port baseline by +2.2 mIoU on ADE20K.
arXiv:2603.22281v2 Announce Type: replace Abstract: Recent progress in latent world models (e.g., V-JEPA2) has shown promising capability in forecasting future world states from video observations. Nevertheless, dense prediction from a short observation window limits temporal context and can bias predictors toward local, low-level extrapolation, making it difficult to capture long-horizon semantics and reducing downstream utility. Vision--language models (VLMs), in contrast, provide strong semantic grounding and general knowledge by reasoning over uniformly sampled frames, but they are not ideal as standalone dense predictors due to compute-driven sparse sampling, a language-output bottleneck that compresses fine-grained interaction states into text-oriented representations, and a data-regime mismatch when adapting to small action-conditioned datasets. We propose a VLM-guided JEPA-style latent world modeling framework that combines dense-frame dynamics modeling with long-horizon semantic guidance via a dual-temporal pathway: a dense JEPA branch for fine-grained motion and interaction cues, and a uniformly sampled VLM \emph{thinker} branch with a larger temporal stride for knowledge-rich guidance. To transfer the VLM's progressive reasoning signals effectively, we introduce a hierarchical pyramid representation extraction module that aggregates multi-layer VLM representations into guidance features compatible with latent prediction. Experiments on hand-manipulation trajectory prediction show that our method outperforms both a strong VLM-only baseline and a JEPA-predictor baseline, and yields more robust long-horizon rollout behavior.
arXiv:2604.03137v2 Announce Type: replace Abstract: We present a molecular extension of our recently proposed Green's function embedding method, interacting-bath dynamical embedding theory (ibDET), for computing charged excitation energies at the $GW$ and EOM-CCSD levels. Starting from atom-centered impurities, we construct bath representations that capture the frequency-dependent entanglement between the impurity and its environment and can be systematically improved via the construction of cluster-specific natural orbitals. Utilizing a $GW$ or coupled-cluster Green's function solver, the self-energy of the full system is assembled from all embedding problems to obtain the interacting Green's function. We show that ibDET provides accurate spectral properties with much reduced cost for a broad range of systems, including conjugated molecules and nanoclusters. Compared with full-system results, the errors in the predicted ionization potentials and electron affinities are around 0.1 eV or smaller, while each embedding problem includes only a small fraction of the total orbital space. This work provides an efficient and scalable framework for computing spectral properties of molecular systems.
arXiv:2606.18023v1 Announce Type: new Abstract: Looped Transformers scale latent computation by repeatedly applying shared blocks, but sequential looping increases latency and KV-cache memory with the loop count. Parallel loop Transformers (PLT) alleviate this cost through cross-loop position offsets (CLP) and shared-KV gated sliding-window attention, making loop count a practical design choice. We therefore study PLT loop-count selection through a gain--cost view: an extra loop may refine representations, but CLP also introduces a positional mismatch at each loop boundary. We instantiate this study by training LoopCoder-v2, a family of 7B PLT coders with different loop counts, from scratch on 18T tokens, followed by matched instruction tuning and evaluation. Empirically, the two-loop variant delivers broad gains over the non-looped baseline across code generation, code reasoning, agentic software engineering, and tool-use benchmarks, improving SWE-bench Verified from 43.0 to 64.4 points and Multi-SWE from 14.0 to 31.0 points. In contrast, variants with three or more loops regress, revealing a strongly non-monotonic loop-count effect. Our diagnostics show that loop 2 provides the main productive refinement, while later loops yield diminishing, oscillatory updates and reduced representational diversity. Because the CLP-induced mismatch remains roughly fixed as refinement gains shrink, the offset cost increasingly dominates. This gain--cost trade-off explains PLT's saturation at two loops and provides diagnostics for loop-count selection.
arXiv:2603.04438v3 Announce Type: replace-cross Abstract: Fully unsupervised deep generative modeling (FU-DGM) offers significant potential for compressively sampled magnetic resonance imaging (CS-MRI) reconstruction. Representative FU-DGM formulations, such as deep image prior (DIP) and implicit neural representation (INR), employ architectural bias to induce a low-dimensional manifold in the image space that aligns with the forward observation. However, as the underlying inverse system is highly ill-posed, prolonged iterative fitting in FU-DGM typically leads to poor efficiency and noise amplification. In this paper, guided by the cognitive principle of easy-to-hard learning, we propose CogGen, an FU-DGM framework that reformulates CS-MRI reconstruction as a staged inversion problem. Specifically, CogGen implements an self-paced curriculum learning (SPCL)-driven progressive scheduling strategy through an MRI-aware dual-threshold weighting criterion, which adaptively regulates k-space measurement participation. The data-consistency residual thresholding evaluates the fitting reliability of the current generator, while the k-space radius thresholding controls stage-wise measurement exposure, thereby avoiding uniform fitting throughout optimization. Theoretically, our analysis shows that, when early stages favor easy-to-fit measurements, CogGen yields a reduced local sufficient-iteration bound and a smaller cumulative noise-amplification bound, explaining the improved convergence behavior and reconstruction fidelity of CogGen within a finite iteration budget. Numerical experiments demonstrate that both CogGen instantiations, CogGen-DIP and CogGen-INR, achieve superior performance over prevailing CS-MRI reconstruction techniques, including unsupervised and supervised pipelines.
arXiv:2507.14632v4 Announce Type: replace Abstract: The rapid advancement of generative AI has substantially improved image and video synthesis, amplifying the risk of multimodal visual misinformation. Recent MLLMs have shown promise for transparent AI-generated content detection through reasoning and explanation, yet existing approaches largely treat image and video forensics as isolated tasks, leaving cross-modal synergies underexplored. To address this, we present \textbf{BusterX++}, a unified MLLM for joint image and video detection with interpretable reasoning. We also introduce \textbf{GenBuster-Bench++}, a meticulously curated, difficulty-aligned benchmark containing balanced image and video samples spanning recent generation models and diverse real-world scenarios. Using this controlled setting, we revisit the widely adopted $SFT \rightarrow RL$ post-training paradigm. Notably, our findings demonstrate that a single-stage, pure RL strategy driven strictly by sparse outcome rewards consistently matches or surpasses a strong SFT+RL baseline across both unified and single-modality settings. Our key insight reveals that SFT imposes lower policy entropy, which restricts the policy search space and dampens exploratory freedom. In contrast, single-stage pure RL maintains higher policy entropy throughout training, effectively unlocking the spontaneous emergence of cross-modal capability transfer between image and video forensics. Extensive experiments demonstrate that BusterX++ achieves state-of-the-art performance, highlighting the powerful potential of RL for unified cross-modal visual reasoning.
arXiv:2509.00064v2 Announce Type: replace Abstract: Robotic practices on the construction site emerge as an attention-attracting manner owing to their capability of tackling complex challenges, especially in the rebar-involved scenarios. Most of existing products and research are mainly focused on the collection of large amounts of data with model training demands. To fulfill this gap, we propose OpenTie, a 3D training-free rebar tying framework utilizing a RGB-to-point-cloud generation and an open-vocabulary rebar detection on the real-world test. We implement the OpenTie via a robotic arm with a binocular camera and guarantee a high accuracy by applying the prompt-based object detection method on the image filtered by our proposed post-processing procedure for the image-to-point-cloud generation framework. Our pipeline requires no training efforts and outperforms the training-based object detection, i.e., YOLO-based method, with the verification on the real-world sequential rebar tying test. The system is flexible for horizontal and vertical rebar tying tasks and holds the potential application to the real construction site with possibility of commercialization.
arXiv:2606.18075v1 Announce Type: new Abstract: Retrieval-Augmented Generation (RAG) has emerged as a paradigm for enhancing large language models (LLMs) with external knowledge, yet existing graph-based methods face a fundamental limitation: entity-centric and chunk-centric approaches operate on representations anchored to original text without true knowledge fusion. While entity-centric methods connect logically related content and chunk-centric methods preserve context, both retrieve information separately through similarity search, missing emergent understanding from their synthesis. In this paper, we propose HyGRAG, a hierarchical graph RAG framework that transcends source documents by addressing three core challenges: constructing summaries that genuinely integrate contextual and relational information, leveraging these synthesized representations to access emergent knowledge during retrieval, and efficiently updating hierarchical structures for dynamic corpora. Specifically, we design hierarchical index structures over hybrid graphs with both chunk and entity nodes, then iteratively cluster them and generate LLM-based summaries. Then, we design context and relation-aware retrieval that searches across all abstraction levels while expanding through community membership. Moreover, we enable dynamic knowledge update through attachment-based algorithms with only local re-summarization. Experimental results show that HyGRAG improves the average accuracy of multi-hop reasoning tasks by 9.7%, while maintaining reasonable efficiency.
arXiv:2606.18080v1 Announce Type: new Abstract: Gradient descent in deep learning may operate at the edge of stability (EoS), a regime in which the largest eigenvalue of the loss Hessian hovers near the stability threshold $2/\eta$, where $\eta$ is the learning rate. Classical analysis tools such as gradient flow and the descent lemma do not apply here, motivating the search for a continuous-time model valid at EoS. We propose Edge Flow, a system of three coupled ordinary differential equations that provides a tractable, faithful, and predictive model of gradient descent dynamics at EoS. Edge Flow decomposes the dynamics into a center, an oscillation direction, and an oscillation magnitude. The center follows a modified gradient flow on a symmetrized loss; the direction tracks a top eigenvector of the Hessian via Rayleigh quotient dynamics; and the magnitude grows or decays exponentially depending on whether the sharpness exceeds or falls below the threshold $2/\eta$. Crucially, sharpness stabilization emerges from the coupled dynamics via a self-stabilization feedback loop. Discretizing Edge Flow only requires two gradient evaluations and one Hessian--vector product at each iteration. We demonstrate empirically that Edge Flow tracks the dynamics of gradient descent at least as faithfully as previously proposed continuous-time EoS models, while in addition resolving the oscillation of the sharpness at the onset of EoS, and that it provides a principled framework for understanding and mitigating instabilities in this regime.
arXiv:2606.17692v1 Announce Type: new Abstract: Accurate short-term electricity load forecasting is critical for the reliable and economic operation of modern power systems, under non-stationarity arising from weather variability, calendar effects, and evolving consumption patterns. While deep learning models such as LSTMs and Transformers show promising performance, most existing studies focus on direct absolute load prediction without explicitly addressing target non-stationarity. Motivated by classical time-series differencing techniques in ARIMA models, this paper investigates a delta-based target reformulation for short-term electricity load forecasting using deep learning. Instead of directly predicting absolute load values, the proposed formulation trains models to predict the change in load between consecutive time steps, with final forecasts reconstructed using the last observed load. This aims to stabilize the learning target and reduce forecasting difficulty. Using multi-year, hourly real-world electricity load data from India, augmented with meteorological variables from the NASA POWER project and calendar features, this study evaluates LSTM and Transformer models under both formulations, benchmarking them against LightGBM. Experiments are conducted for hour-ahead and day-ahead horizons, assessing performance via Mean Absolute Error (MAE) and Mean Absolute Percentage Error (MAPE). Results show that delta-based reformulation consistently improves forecasting accuracy for hour-ahead prediction across all evaluated models, yielding MAPE reductions of over 50% compared to absolute formulations. For day-ahead forecasting, delta targets specifically benefit deep sequence models (LSTM and Transformer), while LightGBM remains competitive under the absolute formulation. These findings indicate that while delta reformulation is a powerful inductive bias for neural networks, its efficacy is model- and horizon-dependent.
arXiv:2606.18111v1 Announce Type: new Abstract: Fairness is an important aspect of decision-making in multi-objective reinforcement learning (MORL), where policies must ensure both optimality and equity across multiple, potentially conflicting objectives. While single-policy MORL methods can learn fair policies for fixed user preferences using welfare functions such as the generalized Gini welfare function (GGF), they fail to provide the diverse set of policies necessary for dynamic or unknown user preferences. To address this limitation, we formalize the fair optimization problem in multi-policy MORL, where the goal is to learn a set of Pareto-optimal policies that ensure fairness across all possible user preferences. Our key technical contributions are threefold: (1) We show that for concave, piecewise-linear welfare functions (e.g., GGF), fair policies remain in the convex coverage set (CCS), which is an approximated Pareto front for linear scalarization. (2) We demonstrate that non-stationary policies, augmented with accrued reward histories, and stochastic policies improve fairness by dynamically adapting to historical inequities. (3) We propose three novel algorithms, which include integrating GGF with multi-policy multi-objective Q-Learning (MOQL), state-augmented multi-policy MOQL for learning non-statoinary policies, and its novel extension for learning stochastic policies. We evaluate our algorithms across various domains and compare our methods against the state-of-the-art MORL baselines. The empirical results show that our methods learn a set of fair policies that accommodate different user preferences.
arXiv:2605.20214v2 Announce Type: replace Abstract: Based on the fractional entropy originating from fractional quantum mechanics, the fractional holographic dark energy (FHDE) model has been proposed. In this paper, we consider an interaction between the pressureless matter and FHDE and analyze three different interacting FHDE models. Combining the latest observational data including SNIa, OHD, BAO, and CMB, we estimate the model parameters and find that the interaction forms $Q=\gamma H \rho_{de}$ and $Q=\beta H \rho_{m}+\gamma H \rho_{de}$ show some preference from the observational data. Using phase space analysis, we further find that only interacting FHDE model with $Q=\beta H \rho_{m}+\gamma H \rho_{de}$ can describe the full evolutionary history of the universe. The statefinder diagnostic pair reveals that this model deviates from the $\Lambda$CDM model but converges to the $\Lambda$CDM fixed point and the de Sitter expansion fixed point in the future. Finally, we analyze the evolution of cosmological parameters and demonstrate that this model can drive the late time acceleration of the universe.
arXiv:2606.15817v2 Announce Type: replace Abstract: Two Scratch programs can be syntactically far apart-renamed variables, split scripts, extracted custom blocks, or reordered initialization-and still behave identically; a one-block edit, such as replacing a blocking broadcast with an asynchronous one, can create divergences visible only under specific schedules. Deciding behavioral equivalence is central to automated feedback, grading support, and repair validation, yet tree differencing is too strict and single-run dynamic comparison is unsound for concurrent, random, and timing-dependent behavior. We observe that equivalence for block-based programs is lens-parametric: final state, frame traces, monitors, event causality, and debug traces induce different observation relations. ScratchLens makes this explicit through a taxonomy of causal divergence phenomena and observation lenses. It compiles Scratch projects into a causal IR of typed resources and semantic transactions, canonicalizes renamings, guards, and procedure bodies, quotients same-trigger concurrency with Mazurkiewicz trace normal forms, separates program order from races, and handles residual frontiers through SMT obligations and VM-backed counterexample-guided refinement. Conclusive verdicts carry evidence: equivalence by bijection and trace quotient, difference by a typed witness, and unresolved cases remain unknown. On a VM-witnessed mutation corpus from real Scratch projects, ScratchLens decides all 444 validated pairs and makes 0/158 false-equivalence claims on witnessed-different pairs under strict scoring. Structural, dynamic-only, and LLM baselines fail on the classes predicted by the taxonomy; ablations quantify the contribution of partial-order reduction and lens parametricity; and targeted scenarios expose ambiguous-mutant divergences missed by random testing.
arXiv:2308.00651v5 Announce Type: replace-cross Abstract: Markov categories have recently turned out to be a powerful high-level framework for probability and statistics. They accommodate purely categorical definitions of notions like conditional probability and almost sure equality, as well as proofs of fundamental results such as the Hewitt--Savage 0/1 Law, the de Finetti Theorem and the Ergodic Decomposition Theorem. In this work, we develop additional relevant notions from probability theory in the setting of Markov categories. This comprises improved versions of previously introduced definitions of absolute continuity and supports, as well as a detailed study of idempotents and idempotent splitting in Markov categories. Our main result on idempotent splitting is that every idempotent measurable Markov kernel between standard Borel spaces splits across another standard Borel space, and we derive this as an instance of a general categorical criterion for idempotent splitting in Markov categories.
arXiv:2606.18124v1 Announce Type: new Abstract: Modern conversational AI systems frequently rely on user metadata to localize responses, yet the unintended regional biases introduced by this hidden context remain poorly understood. In this work, we evaluate location leakage: the phenomenon where a model generates geographic references despite receiving a geographically neutral user prompt. Across both creative writing and open-ended Q&A prompts, even state-of-the-art LLMs systematically favor region-specific outputs when exposed to location metadata, with leakage spiking by up to 793 times above baseline (e.g., from 0.04% to 31.7% for Llama 3.1-8B, and 21.3% and 8.8% for Qwen3-8B and Claude Sonnet 4.6, respectively). Our analysis further shows a novel structural conditioning effect: replacing the injected location with the placeholder "Unknown" still elevates leakage by up to 72 times above baseline, demonstrating that the user profile frame itself, independent of any geographic content, acts as a generative conditioning signal.
arXiv:2606.18185v1 Announce Type: new Abstract: We propose a machine-learning framework for detecting whether a given matrix is a perturbation of a matrix with a large Jordan block. The proposed model achieves high classification accuracy on synthetically generated, robustly perturbed data and outperforms a classical numerical baseline. Moreover, we demonstrate that the learned model generalizes to several classes of matrices not seen during training. These results suggest that the architecture captures structural properties associated with matrix defectiveness.
arXiv:2508.02721v2 Announce Type: replace Abstract: While powerful, the inherent non-determinism of large language model (LLM) agents limits their application in structured operational environments where procedural fidelity and predictable execution are strict requirements. This limitation stems from current architectures that conflate probabilistic, high-level planning with low-level action execution within a single generative process. To address this, we introduce the \textsc{Source Code Agent} framework, a new paradigm built on the ``Blueprint First, Model Second'' philosophy that decouples workflow logic from the generative model. An expert-defined operational procedure is first codified into a source code-based Execution Blueprint, which is then executed by a deterministic engine. The LLM is strategically invoked as a specialized tool to handle bounded, complex sub-tasks within the workflow, but never to decide the workflow's path. We evaluate on the TravelPlanner benchmark for constraint-aware travel planning. The \textsc{Source Code Agent} achieves a 35.56\% final pass rate, a 97.6\% improvement over the state-of-the-art ATLAS baseline (18.00\%) on the same Claude-Sonnet-4 backbone. Critically, it reduces constraint violations by 96.0\% (11 vs 275) while improving execution efficiency by 27.1\% (10.2$\pm$0.7 steps vs 14.0). Two production incident-diagnosis deployments and additional results on ScienceWorld and ALFWorld confirm that the architecture transfers beyond travel planning to procedurally well-defined, constraint-intensive workflows. Our work enables the verifiable and reliable deployment of autonomous agents in applications governed by strict procedural logic.
arXiv:2606.17117v1 Announce Type: cross Abstract: Over-the-air federated learning (FL) leverages the superposition property of multiple-access channels to enable communication-efficient distributed model training. Existing integrated sensing, communication, and computation (ISCC)-enabled over-the-air FL systems typically require dedicated resources for the sensing module, inevitably compromising FL performance due to resource competition. In this paper, we propose a sensing-native over-the-air FL framework that explores built-in distributed wireless sensing capability with zero overhead per model aggregation. Specifically, the high-dimensional local gradient signals possessing favorable autocorrelation property are concurrently leveraged for target distance estimation, while the gradient statistics already required for over-the-air FL serve as a ready-made gateway to deliver locally-sensed results to the edge server for cooperative localization. To combat inter-device interference, channel fading, and communication noise, we put forth a robust trilateration-based target positioning method building upon an efficient matched-filtering-based distance estimation. Then, by explicitly characterizing the impact of imperfect model aggregation and noisy gradient-statistics transmission on the sensing-native over-the-air FL convergence, we develop a statistics-aware communication-learning co-design approach. We first derive the closed-form optimal power budgets allocated to local gradients and their statistics, based on which an efficient successive convex approximation method is proposed for receiver beamforming optimization. Simulation results show that the proposed framework simultaneously achieves superior learning and sensing performance compared to representative baselines.
arXiv:2509.26633v3 Announce Type: replace Abstract: A dominant paradigm for teaching humanoid robots complex skills is to retarget human motions as kinematic references to train reinforcement learning (RL) policies. However, existing retargeting pipelines often struggle with the significant embodiment gap between humans and robots, producing physically implausible artifacts like foot-skating and penetration. More importantly, common retargeting methods neglect the rich human-object and human-environment interactions essential for expressive locomotion and loco-manipulation. To address this, we introduce OmniRetarget, an interaction-preserving data generation engine based on an interaction mesh that explicitly models and preserves the crucial spatial and contact relationships between an agent, the terrain, and manipulated objects. By minimizing the Laplacian deformation between the human and robot meshes while enforcing kinematic constraints, OmniRetarget generates kinematically feasible trajectories. Moreover, preserving task-relevant interactions enables efficient data augmentation, from a single demonstration to different robot embodiments, terrains, and object configurations. We comprehensively evaluate OmniRetarget by retargeting motions from OMOMO, LAFAN1, and our in-house MoCap datasets, generating over 8-hour trajectories that achieve better kinematic constraint satisfaction and contact preservation than widely used baselines. Such high-quality data enables proprioceptive RL policies to successfully execute long-horizon (up to 30 seconds) parkour and loco-manipulation skills on a Unitree G1 humanoid, trained with only 5 reward terms and simple domain randomization shared by all tasks, without any learning curriculum.