arXiv:2604.25496v2 Announce Type: replace Abstract: Offline zero-shot reinforcement learning (RL) aims to learn agents that optimize unseen reward functions without additional environment interaction. The standard approach to this problem trains task-conditioned policies by sampling task vectors that define linear reward functions over learned state representations. In most existing algorithms, these task vectors are randomly sampled, implicitly assuming this adequately captures the structure of the task space. We argue that doing so leads to suboptimal zero-shot generalization. To address this limitation, we propose extracting task vectors directly from the offline dataset and using them to define the task distribution used for policy training. We introduce a simple and general reward function extraction procedure that integrates into existing offline zero-shot RL algorithms. Across multiple benchmark environments and baselines, our approach improves zero-shot performance by an average of 20%, highlighting the importance of principled task sampling in offline zero-shot RL.
Science Journals
arXiv:2605.04926v2 Announce Type: replace Abstract: Promotional language has been increasingly used to aid the communication of innovative ideas in science. Yet, less is known about its role in the context of technological innovation. Here, we use a validated and domain-diagnosed lexicon of 135 promotional words to study the association between promotional language and patent evaluation outcomes among 2.7 million USPTO patent applications. Our large-scale study reveals three unexpected findings. First, in contrast to scientific evaluation, we find that a higher frequency of promotional words is negatively associated with the probability of an application being (i) granted a patent, (ii) transferred ownership, and (iii) successfully appealed. This promotional penalty holds even after accounting for a range of confounding factors and is largely robust across different technological areas. Among matched samples, the difference in the success rate between the lowest and highest promotional density quintile is 5.5, 5.9, and 5.3 percentage points for patentability, transferability, and rejection reversal. Second, contrary to institutional skepticism, we show that promotional language is not a mask of weak technology, but objectively reflects the degree of combinatorial novelty and future citation impact. Third, digging into the mechanisms, we find that the tolerance to promotional framing is strongly moderated by human factors, with men and experienced examiners showing a higher acceptance of promotional narratives than women and novice examiners. By revealing an emerging paradox in the patent system, our study offers theoretical and practical implications for improving patent evaluation through more objective scrutiny of linguistic patterns in patent filings.
arXiv:2606.00260v3 Announce Type: replace Abstract: Human Activity Recognition (HAR) from ambient sensors enables smart-home applications such as health monitoring and assisted living. In realistic deployments, however, sensor events arrive as a continuous stream and activity boundaries are unknown. Sliding-window inference therefore produces many windows that straddle transitions and contain mixed activities, creating boundary contamination that violates the pre-segmented instance assumption used by most benchmarks and models. Moreover, many pipelines under-use spatial context by treating sensor IDs as independent tokens. We present LastAct, a trajectory-centric framework for streaming smart-home HAR that targets the most recent activity under mixed windows while explicitly modeling spatial structure. LastAct projects sensor events onto the home floorplan to form a layout-aligned trajectory image sequence that preserves spatial continuity. A lightweight gate identifies contaminated windows, and a boundary localizer estimates the last transition to enable boundary-guided masking that emphasizes post-boundary evidence and suppresses stale context. For efficiency, we reuse a precomputed layout-aligned template cache to avoid repeated rendering. Empirically, across four public smart-home datasets under near-realistic mixed-activity protocols, LastAct achieves competitive or superior performance on pure windows and yields substantial Macro-F1 gains on cross/mixed windows, demonstrating improved robustness under near-realistic sliding-window regimes.
arXiv:2606.08674v3 Announce Type: replace Abstract: Video generation for biological behavior requires more than visually plausible motion: the duration of an action is itself a semantic property. Existing models usually rely on fixed temporal windows, external continuation, or prompt-driven stories, so length is specified externally rather than learned from behavior. To address this gap, we propose BioVid, a data-driven autoregressive framework for adaptive-length biological behavior generation. BioVid uses a 2D-encode/3D-decode tokenizer: a two-dimensional FSQ-R3GAN encoder converts each frame into discrete visual tokens, preserving single-frame information suited for next-token prediction and EOS-based termination, while a temporally inflated and video-finetuned three-dimensional decoder reconstructs generated tokens with temporal context to reduce flickering. A causal Transformer then models the frame-wise token sequence and, conditioned only on the first frame, stops generation when it emits an End-of-Sequence token, allowing duration to emerge from the learned behavior distribution. We evaluate BioVid on the A001 drinking action from NTU RGB+D. On 94 held-out clips, BioVid achieves a Wasserstein-1 distance of 1.24 frames from the real duration distribution. In comparison, fixed-length baselines yield distances of approximately 6-7 frames even when configured to the available length closest to the dataset mean, and approximately 15 frames when using the conventional 16-frame generation length. These results demonstrate the ability of BioVid to learn and reproduce the intrinsic duration distribution of biological behavior.
arXiv:2606.21051v2 Announce Type: replace Abstract: Nonlocal metasurfaces enable all-optical processing of spatial information in optical fields. Here, we demonstrate a topology-optimised metasurface that switches between phase-contrast and brightfield imaging modalities via polarisation control, eliminating the need to reposition optical components or use computational techniques to image transparent samples. Specifically, for one polarisation state, an asymmetric transfer function about normal incidence performs a first order derivative on the phase, producing pseudo-3D phase-contrast images of transparent biological samples while the orthogonal state returns the result of the identity operator. This work extends inverse-design methods to reconfigurable phase-contrast microscopy and quantitative analogue optical computation with applications in biological imaging, medical diagnostics, and materials characterisation.
arXiv:2604.26748v2 Announce Type: replace Abstract: Robust Markov decision processes (RMDPs) extend standard Markov decision processes (MDPs) to account for uncertainty in the transition probabilities. RMDPs have an uncertainty set that defines a set of possible transition functions, each of which induces a standard MDP. The natural objective in an RMDP is to optimize the discounted cumulative reward under the worst-case transition function in the uncertainty set. We study the complexity of the associated threshold problem for RMDPs with polytopic uncertainty sets in halfspace representation. Previous results focused on approximating the optimum or restricted attention to specific subclasses of RMDPs, such as interval MDPs or $L_\infty$-RMDPs. Our contributions are threefold: (1) For (s,a)-rectangular RMDPs, we prove that robust policy evaluation is in P via robust linear programming, and that the threshold problem is in NP. As a corollary, robust policy iteration is a polynomial-time algorithm for these RMDPs when the discount factor is fixed. (2) For $s$-rectangular RMDPs, we show that the threshold problem is in PSPACE via the first-order theory of the reals. (3) We establish lower bounds by reducing both parity games and bisimulation metrics between MDP states to the RMDP threshold problem. A polynomial-time algorithm for the threshold problem would resolve the long-standing open question of whether parity games can be solved in polynomial time. The reduction from bisimulation metrics also yields a practical benefit: it allows us to apply robust policy iteration as a more efficient alternative to the standard fixed-point iteration, as our empirical evaluation demonstrates.
arXiv:2605.00327v4 Announce Type: replace Abstract: In large language model (LLM)-based recommendation systems, direct preference optimization (DPO) effectively aligns recommendations with user preferences, requiring multi-negative objective functions to leverage abundant implicit-feedback negatives and sharpen preference boundaries. However, our empirical analyses reveal a counterintuitive phenomenon, preference optimization collapse, where increasing the number of negative samples can lead to performance degradation despite a continuously decreasing training loss. We further theoretically demonstrate that this collapse arises from gradient suppression, caused by the dominance of easily discriminable negatives over boundary-critical negatives that truly define user preference boundaries. As a result, boundary-relevant signals are under-optimized, weakening the model's decision boundary. Motivated by these observations, we propose DynamicPO (Dynamic Preference Optimization), a lightweight and plug-and-play framework comprising two adaptive mechanisms: Dynamic Boundary Negative Selection, which identifies and prioritizes informative negatives near the model's decision boundary, and Dual-Margin Dynamic beta Adjustment, which calibrates optimization strength per sample according to boundary ambiguity. Extensive experiments on three public datasets show that DynamicPO effectively prevents optimization collapse and improves recommendation accuracy on multi-negative preference optimization methods, with negligible computational overhead. Our code and datasets are available at https://github.com/xingyuHuxingyu/DynamicPO.
arXiv:2606.25950v1 Announce Type: new Abstract: In our increasingly interconnected world, good IT security practices are necessary to prevent vulnerabilities and data breaches. Providing security contacts, e.g., via Coordinated Vulnerability Disclosure (CVD) programs or security.txt files, is an important practice for businesses to facilitate vulnerability reporting by external parties. As part of a longitudinal study, we analyzed the adoption of, as well as the challenges and experiences with, CVD programs among the 40 companies listed on Germany's DAX (the country's primary stock market index). In addition to monitoring publicly available information about their CVD programs, we sent out questionnaires via email and postal mail in 2023 and 2025, and received answers from 20\% of the companies. The adoption rates show a significant increase from 50\% (2023) to over 90\% (2025), with ten new CVD programs and 25 new security.txt files now available. The survey answers reveal that, for example, legal obligations (e.g., NIS2 and CRA) drive the adoption of CVD practices, but a lack of (human) resources and varying report quality are considered drawbacks. As the first study to survey 40 German stock market index (DAX) companies on their CVD practices, our results can help foster the adoption and understanding of security programs among SMEs and other companies, and provide policymakers with insights into practical challenges and industry experiences.
arXiv:2605.01708v3 Announce Type: replace Abstract: Contemporary systems serving large language models (LLMs) have adopted prefill-decode disaggregation to load-balance between the compute-bound prefill phase and the memory-bound decode phase. Under this design, prefill workers generate a KV cache that must be transferred to decode workers before generation can begin. With these workers residing on different physical systems, this transfer becomes a significant bottleneck to serving LLMs at scale, especially for long-input and agentic workloads. Existing lossless codecs are unsuitable here as they primarily target offline weight compression, run on CPUs, or use variable-length coding whose compression cannot keep up with KV production during prefill. We introduce SplitZip, a GPU-friendly lossless compressor for KV cache transfer that preserves KV tensors bitwise and integrates into existing serving frameworks without modifying model execution. SplitZip exploits redundancy in floating-point exponents of KV activations, encoding frequent exponent values with fixed-length codes and routing rare exponents through a sparse escape stream of (position, value). A calibrated top-16 exponent codebook eliminates online histogramming, while the regular dense path and sparse escape correction make both encoding and decoding efficient on GPUs. On real BF16 activation tensors, SplitZip achieves $613.3$ GB/s compression throughput and $2181.8$ GB/s decompression throughput, outperforming prior lossless compressors on the critical codec path. End-to-end transfer experiments show up to $1.32\times$ speedup for BF16 KV cache transfer, $1.30\times$ speedup for TTFT, and $1.23\times$ increase in Request Throughput. The same approach extends to FP8 KV caches, providing up to $1.14\times$ compression over native E5M2. Code is available at https://github.com/Intelligent-Microsystems-Lab/SplitZip
arXiv:2605.07412v2 Announce Type: replace Abstract: Although Global Navigation Satellite Systems (GNSS) provide a general solution for bike tracking outdoors, there still exist complex riding environments where only inertial navigation systems work, such as urban canyons. Despite decades of research, localization using only low-cost inertial sensors still faces challenges such as cumulative drifts and poor robustness caused by filtering methods. Furthermore, sensors such as visual and LiDAR could provide reliable measurements, but they are not suitable for large-scale deployment. In this paper, we propose an inertial tracking framework that integrates bicycle mechanical constraints with a mixture-of-experts model. Specifically, we leverage multiple expert modules to capture shared representations and weight them through the gating mechanism, thus improving multi-task learning performance and enabling uncertainty-aware trajectory estimation. Furthermore, based on the mechanical transmission between the pedal and the rear wheel of a bike, we explore the intrinsic relationship between the rider's periodic pedalling behaviors and acceleration variations, and convert such patterns into bike's wheel speed for dynamic calibration. Experiments with real-world riding data from shared bikes of the DiDi ride-hailing platform demonstrate that our system improves the accuracy of baselines by at least 12%, with wheel speed errors below 0.5 m/s at 95-percentile.
arXiv:2605.13076v3 Announce Type: replace Abstract: The LLM-based generation of machine-readable outputs such as JSON has attracted significant attention for integration with external systems. However, existing approaches cannot strictly enforce the maximum number of tokens to be generated, leading to infinite generation or truncated outputs that cause a system malfunction. To address this limitation, we propose TruncProof, a novel grammar-constrained generation method that enables LLMs to produce grammatically valid JSONs while adhering to a predefined token limit. By leveraging the properties of LL(1) parsers, TruncProof efficiently approximates the minimum number of tokens required to complete a grammatically valid output at each decoding step. Experiments on the Text-to-JSON instruction tasks demonstrate that TruncProof successfully generates syntactically correct outputs even under strict token constraints. Furthermore, we show that TruncProof can be effectively combined with advanced decoding strategies, resulting in outputs that are not only grammatically valid but also semantically accurate. The source code is public at https://github.com/Yosshi999/TruncProof
arXiv:2606.25953v1 Announce Type: new Abstract: Existing object-aware SLAM systems force a trade-off between real-time performance, multi-class support, and the generation of high-fidelity, semantically coherent object models. To address this trade-off, we present DSP-SLAM++, which extends the DSP-SLAM framework with an asynchronous mapping pipeline for real-time performance and dedicated sensor fusion adaptations for a monocular fisheye-LiDAR suite. Experiments demonstrate that our system generates fine-grained, geometrically-complete shapes for multiple object classes while eliminating severe mapping thread bottlenecks by reducing maximum object processing latency by up to 70\% compared to the state-of-the-art baseline, enabling robust, real-time performance on a challenging 25 Hz multi-class datasets. This work makes high-fidelity, multi-class object SLAM more practical for real-world applications like autonomous driving and robotic manipulation by enabling its use on platforms with common fisheye-LiDAR sensor setups. The open-source code is available at: [github.com/AUBVRL/DSP-SLAMpp].
arXiv:2605.16314v2 Announce Type: replace Abstract: We present a phase-based formulation of special relativity in which the kinematical structure of the theory is reconstructed from the requirement of phase coherence of localized wave states. Starting from the assumption that physical propagation is associated with surfaces of constant phase and that matter admits an intrinsic rest-frame oscillation, we show that the relativistic relations for time dilation, energy, and momentum follow from the invariant accumulation of phase along particle trajectories. In this framework, proper time is identified operationally as the phase count of an internal clock, and the Minkowski interval arises as the quadratic form consistent with phase invariance across inertial observers. The relation between mass and rest-frame frequency emerges naturally, providing a unified interpretation of relativistic kinematics in wave-mechanical terms. This formulation does not introduce new dynamics but offers a coherent structural perspective that bridges relativistic kinematics and wave propagation.
arXiv:2605.22121v2 Announce Type: replace Abstract: Magnetic resonance imaging (MRI) is highly susceptible to patient motion due to its relatively long acquisition times and the fact that data are acquired sequentially in k-space. Even small patient movements introduce phase inconsistencies across measurements, leading to severe artifacts such as blurring, ghosting, and geometric distortions that can compromise diagnostic quality. Retrospective motion compensation remains challenging, particularly in accelerated acquisitions, due to the ill-posed nature of the joint reconstruction and motion estimation problem. In this work, we propose a unified Bayesian framework for motion-compensated 3D MRI that jointly estimates the anatomical image, rigid-body motion parameters, and coil sensitivity maps directly from motion-corrupted k-space data. Our approach integrates pretrained 3D complex-valued score-based diffusion models as expressive anatomical image priors within a physics-based forward model. Inference is performed by alternating diffusion posterior image updates with efficient proximal optimization steps for motion and coil sensitivity estimation, enabling fully unsupervised reconstruction without the need for paired motion-free training data. Experiments on simulated and real-motion brain MRI datasets demonstrate that the proposed method achieves improved image quality and motion robustness compared to state-of-the-art classical and learning-based motion correction techniques, particularly in the presence of severe motion and high acceleration.
arXiv:2605.26309v3 Announce Type: replace Abstract: The rapid proliferation of visual content raises fundamental questions about how different visual formats and features shape perceived credibility. Drawing on processing fluency theory, this research examines how visuals shape credibility judgments. We focus on three popular formats-photos, infographics, and data visualizations-comparing them to text-only posts, and test how two visual features, aesthetic appeal and production quality, influence credibility through processing fluency as a mediating mechanism. Through a preregistered experiment with 1200 US participants, we found that visual posts are generally perceived as more credible than text-only posts but this credibility advantage only applies to photos and infographics, not to data visualizations. Aesthetic appeal increases perceived credibility, partially mediated by processing fluency, while production quality had no significant effect on credibility across formats. These findings differentiate visual formats, advance conceptualizations of visual features, and identify processing fluency as a key mechanism for theorizing credibility across multimodal contexts.
arXiv:2606.25962v1 Announce Type: new Abstract: Modern stereo-capable smartphones enable immersive XR content capture. However, hardware heterogeneity across camera modules often causes severe asymmetric blur artifacts. Existing methods and benchmarks largely assume homogeneous stereo setups and therefore do not explicitly address such asymmetric degradation. To bridge this gap, we present a dedicated framework for heterogeneous stereo deblurring. First, we introduce the heterogeneous stereo deblurring (HSD) dataset, constructed from real smartphone stereo captures via multi-frame integration. Second, we propose physically- and epipolar-constrained cross attention (PECA), a lightweight module that restricts cross-view matching to an epipolar search window bounded by a optics-derived disparity upper bound. By enforcing physically valid disparity constraints, PECA enables efficient and reliable cross-view feature fusion. Moreover, our confidence-weighted attention with residual fusion emphasizes cross-guided deblurring when correspondences are reliable, while naturally falling back to self-deblurring in occluded or unreliable regions. PECA is architecture-agnostic and consistently improves CNN-, Transformer-, and NAFNet-based baselines. Extensive experiments on HSD show that PECA-enhanced models achieve improved restoration performance with favorable efficiency.
arXiv:2606.09457v2 Announce Type: replace Abstract: Embodied policies typically map current observations directly to actions, leaving candidate-action consequences implicit. World models provide predictive supervision, representations, or external simulation, but rarely let a policy inspect the imagined consequence of its own proposal before acting. We introduce $\omega$-EVA, a latent interactive world model that realizes an Envision--Verify--Act loop for embodied action generation. Its three-stage framework learns action-conditioned latent dynamics, trains a language-conditioned flow policy on dynamics-aware visual representations, and feeds the policy's proposal back through the world model. A tri-branch refiner jointly reasons over the current state, proposal-conditioned future, and proposed action to produce the final action chunk. Because consequence reasoning remains in latent feature space, $\omega$-EVA avoids generating future videos at inference. Evaluations across diverse single-arm, bimanual, long-horizon, and perturbed simulation settings show that the complete interaction pipeline consistently improves the proposal policy, while latent diagnostics indicate meaningful action-conditioned future structure. With approximately 1.2B parameters and no additional robot-data pretraining, $\omega$-EVA demonstrates a compact and competitive performance--scale--data trade-off, making the world model an active action-feedback module rather than a passive predictor.
arXiv:2606.13837v2 Announce Type: replace Abstract: Quranic Passage Retrieval (PR) could be a challenging task due to the linguistic complexity and the semantic gap between the Modern Standard Arabic (MSA) used in daily queries and the Classical Arabic (CA) of the Holy Quran. These factors hinder conventional retrieval methods. To handle these limitations and improve multi-verse retrieval and filter the zero-answer queries, this paper proposes a four-phase neural architecture designed to enhance retrieval accuracy and contextual understanding. The methodology combines hybrid candidate retrieval using AraColBERT dense indexing and BM25 sparse retrieval, followed by semantic reranking with a CAMeLBERTmix cross-encoder. A confidence gating mechanism is then applied to filter zero-answer queries, and an AraT5-based refinement module for multi-verse aggregation. The system is evaluated on an expanded version of the Quran QA 2022 dataset. Results show improved performance compared to the baseline models, achieving a Recall@10 of 0.7024 and a Mean Average Precision (MAP@10) of 0.4947. While the system exhibits a marginal tradeoff in absolute top-rank precision (MRR = 0.5807) compared to heavily optimised single models, the proposed architecture provides a substantially more comprehensive, reliable, and context aware solution for multi-verse Quranic passage retrieval.
arXiv:2606.17435v2 Announce Type: replace Abstract: Time-series forecasting models remain vulnerable to gradient-based adversarial attacks while existing defense mechanisms typically incur a trade-off in robustness for bounded response and compute cost. The problem is pronounced in Moving Target Defense where maintaining multiple randomized model instances substantially exacerbates the training overhead. In this work, we introduce MorphStrata, a student generation strategy with selective, layer-specific stochastic noise injection that extends the traditional Morphence defense. MorphStrata uses a Transformer backbone as the teacher and perturbs randomly selected architectural blocks to create structured heterogeneity across student models in response to varied data distributions and threat models. We evaluate against vanilla Transformer and Morphence backbones on a suite of benchmarks including the Jena Climate, Electricity Load Diagrams, and Appliances Energy Prediction using FGSM, BIM and PGD attacks across multiple attack strengths. Across datasets and attack regimes, the proposed ensemble maintains comparable adversarial RMSE. Specifically, for high entropy, periodic datasets as in the case of the AEP data, MorphStrata achieves the lowest RMSE across all attacks and perturbation budgets, improving over the static baseline by up to 24.11% and 97.97% under FGSM and BIM respectively at an epsilon value of 0.5 over 30 randomized trials. Targeting the layers to generate MorphStrata students accounts for less than 1% increase in train-times over the Morphence MTD baseline for most of the experiments, while accounting for double digit gains in adversarial RMSE reduction. We also observe a positive correlation between higher pairwise L2 distance (among generated students) and overall defense effectiveness. In summary, MorphStrata maintains adversarial robustness as an MTD defense at marginal cost deltas when compared to existing baselines.
arXiv:2606.25979v1 Announce Type: new Abstract: Runtime monitoring of quantitative signals faces a fundamental trade-off between volatility and over-aggregation: instantaneous observations are noisy, while long-run averages obscure local structure. Localisation measures such as discounted averages offer a principled middle ground, yet remain poorly understood in runtime verification. This paper studies discounted sums from a monitoring perspective, in both deterministic and stochastic settings. We formalize the discounted monitoring problem and show that exact, sound monitoring of discounted sums cannot be achieved with finite memory. To overcome this impossibility, we introduce $\varepsilon$-approximately sound monitoring, deriving explicit bounds on memory and observation requirements. We then extend the framework to stochastic processes via expected discounted sums, defining pointwise and uniform $(\varepsilon,\delta)$-soundness notions, establishing statistical optimality, and proving impossibility beyond a precision threshold. We also formalize the resource complexity of deterministic discounted monitoring via affine register machines and prove a tight worst-case lower bound. Finally, we present a specification language for arithmetic expressions over multiple discounted sums with synchronous and asynchronous semantics, and evaluate our approach on practical scenarios including algorithmic fairness.
arXiv:2606.25980v1 Announce Type: new Abstract: In audiovisual post-production, Foley refers to synchronous sound effects associated with human actions, such as footsteps, cloth rustle, and prop handling, that are recreated to match the on-screen movements and interactions of characters. These sounds are often recorded by professional Foley artists using physical props. This resource-intensive workflow has motivated data-driven research on Foley, including tasks such as classification, retrieval, and generation; however, high-quality annotated Foley datasets for training remain scarce. To address this gap, we present FoleySet, a publicly available Foley dataset of 10,000 audio clips annotated with a two-level Foley taxonomy. This dataset provides a standardized, Creative Commons-licensed resource for data-driven Foley classification, retrieval, and generation.
arXiv:2606.25989v1 Announce Type: new Abstract: Automated classification of marine species from underwater imagery is essential for scalable ocean biodiversity monitoring and conservation policy. Existing approaches struggle with severe domain shift across collection platforms, fine-grained visual similarity between closely related species, and uneven annotation granularity, where many specimens can only be identified to genus or a coarser taxonomic rank. We present a taxonomy-aware deep learning framework that aligns both the training loss and the inference rule with the hierarchical structure of biological classification, combining a taxonomy-weighted loss, minimum-risk Bayesian inference, multi-scale feature encoding, and independent per-rank classification heads. Evaluated on the FathomNet 2025 dataset1 (79 marine classes across seven taxonomic ranks), the system achieves a mean taxonomic distance of 1.581, within 3% of the 1st-place solution (1.535), with the largest gains from metric-aligned inference and simple, decoupled components that generalize better than learned dependencies under distribution shift.
arXiv:2606.26002v1 Announce Type: new Abstract: We present HiReLC, a hierarchical ensemble-reinforcement learning framework for automated joint quantization and structured pruning of deep neural networks. The framework decomposes the compression search across two levels of abstraction: low-level agents (LLAs) operate independently per block, selecting per-kernel configurations over a multi-discrete action space spanning bitwidth, pruning keep-ratio, quantization type, and granularity, while high-level agents (HLAs) coordinate global budget allocation via ensemble voting guided by Fisher Information-based sensitivity estimates. To mitigate the computational cost of policy evaluation, an iterative active learning loop interleaves surrogate-guided RL optimization with post-compression fine-tuning, using a lightweight MLP surrogate to amortize expensive evaluations and a logit-MSE proxy during cold-start. The surrogate is used for reward shaping rather than as a replacement for final post-compression evaluation. The controller is architecture-agnostic by design, with a modular layer abstraction decoupling the RL environment from the underlying network topology. Experiments across Vision Transformer and CNN benchmarks demonstrate effective parameter-storage compression ratios of 5.99 - 6.72$\times$ with a 3.83 % gain in one setting and 0.55 - 5.62 % accuracy drops elsewhere, supporting hierarchical policy decomposition and sensitivity-aware guidance as practical design choices for joint neural network compression.
arXiv:2606.26007v1 Announce Type: new Abstract: Image priors can synthesize target conditions for 3D Gaussian street scenes, but independently edited views do not define a coherent 3D target. Direct fitting can propagate view-specific noise, while existing pipelines do not jointly handle imperfect sparse anchors and standard-rasterizer deployment. To address this gap, teacher-relative appearance residual distillation is introduced for appearance baking. A structured space for frequency decomposition, confidence estimation, and primitive-level lifting is formed by residuals between teacher anchors and original renders. The direct optimization signal is supplied by renderer-space matching, while primitive assignment is regularized by support-aware Gaussian-space aggregation. Supported detail is admitted and unsupported noise is suppressed through confidence-gated coarse-to-fine optimization, after which all residuals are baked into fixed-geometry spherical-harmonic coefficients. The teacher and auxiliary training modules are discarded at inference. Evaluation across Waymo street assets, Tanks and Temples scenes, and multiple target conditions shows a favorable overall balance of target alignment, content preservation, artifact suppression, and cross-view consistency over editing-based baselines. Ablations confirm the effectiveness of the main components. Code will be released at https://github.com/Cagares/Baking-for-3D-Gaussian.
arXiv:2606.26010v1 Announce Type: new Abstract: Robust and accurate odometry estimation is essential in modern robotics. In environments characterized by highly dynamic motion and sensor noise, odometry estimation becomes increasingly challenging. Autonomous racing combines both factors in an unstructured setting, where minimizing odometry latency is essential for stable closed-loop control. This paper introduces FAR-LIO, a highly optimized CUDA-accelerated LiDAR-inertial odometry framework developed for Fast, Accurate, and Robust performance. Our system leverages a novel CUDA-based voxel hashmap to enable parallelized nearest-neighbor search and efficient map updates. We employ a sparsity-aware Generalized Iterative Closest Point algorithm with adaptive thresholding on top of the CUDA-based voxel hashmap with adaptive density to achieve low-latency without compromising accuracy. An Extended Kalman Filter serves as a robust backend. It utilizes an upsampling and delay compensation strategy to fuse the LiDAR odometry with high-frequency IMU data, thereby ensuring a robust and smooth odometry output. We evaluate FAR-LIO across four different sensor setups, using both public datasets and data from two autonomous racecars driving at speeds of up to 250 km/h. FAR-LIO achieves an average 6.9% reduction in the positional error and 38.4% lower runtime compared to state-of-the-art baselines on target hardware using a single parameter set. This demonstrates its computational efficiency and broad applicability. To build upon our work, our code is available open-source on https://github.com/TUMFTM/FAR-LIO.