Papers

Developed as a workhorse for classical simulations of quantum algorithms and quantum many-body systems, Tensor Network methods have entered the scientific mainstream in quantum physics. Among various types of tensor networks, Tensor Trains (commonly know as Matrix Product States in the quantum computing community) have already found applications in machine learning. These methods often rely on a powerful linear algebra tool called the Singular Value Decomposition (SVD). Several conditional GAN architectures for image denoising incorporate SVD as a single-cut decomposition step applied to generator feature maps. In this work we introduce TT-Net, which replaces the per-channel SVD denoising block with a two-cut tensor-train decomposition capable of accessing cross-channel information directly, a capability absent from contemporary alternatives. In a controlled comparison differing only in this decomposition mechanism, TT-Net outperforms SVD-Net on PSNR and SSIM across all three noise types tested (Gaussian, motion blur, and salt-and-pepper), supporting the hypothesis that cross-channel access improves denoising quality. Training-dynamics analysis further shows that TT-Net's adversarial loss term consistently saturates to a stagnant state across all three noise types, more so than SVD-Net's, while reconstruction quality continues to improve regardless, raising an open question about the adversarial component's contribution that this work identifies but does not resolve. Furthermore, for Gaussian noise our method outperforms both the EigenGAN and the state of the art Pix2pix method which does not assume any linear algebra decompositions and does not retain any linear algebra information. Our manuscript shows how quantum inspired tools can be used as practical real world feature filters for deep learning applications.

Trustworthy multimodal fusion in clinical settings requires handling incomplete and heterogeneous modality subsets across institutions, where privacy constraints prohibit centralized data sharing. Federated learning (FL) mitigates data-sharing constraints but suffers from client-specific missing modalities, where institutions possess incomplete multimodal subsets, degrading fusion quality and segmentation performance. While FL and weak supervision have been studied separately, their joint use with image-level labels under heterogeneous missing modalities remains unaddressed. We propose MOSAIC, the first modality-agnostic federated framework for weakly supervised binary tumor segmentation under client-specific missing modalities. We introduce a client-specific modality-alignment module that fuses available channels into a shared latent space without prior knowledge of modality identity, a spectral prototype alignment loss that reconciles cross-client distribution shift using compact non-invertible frequency-domain statistics, and a dedicated federated refinement network that denoises the resulting CAM pseudo-labels into accurate masks, breaking the accuracy ceiling of weak supervision. Experiments on three multi-institutional brain tumor benchmarks (FeTS2022, BraTS-MEN, and BraTS-SSA) demonstrate significant improvements over all image, box, and point-supervised baselines, approaching fully supervised accuracy using only image-level labels and reaching 0.84 Dice on FeTS2022. Dynamic new client addition enables previously unseen institutions to join an already-trained federation within 0.01-0.04 Dice without retraining. Code is available at https://github.com/Tarun2201/MOSAIC.

We show that there exists a total function for which there is a polynomial gap between the randomized and the constant-round quantum communication complexity. Previously, such a separation was known only for quantum protocols using polynomially many rounds.

Discovering dynamical systems from trajectory data is a central problem in applied mathematics and engineering. Whilst recent advances in machine learning have led to strong progress in data-driven system identification, much less attention has been given to systems with discontinuous dynamics. These systems are nevertheless highly relevant in applications, including climate dynamics and mechanical systems with friction. In this work, we consider the problem of identifying piecewise-smooth dynamical systems directly from trajectory data. Compared with the smooth setting, this requires recovering the governing equations and detecting the switching hyperplanes that separate different dynamical regimes and characterising their behaviour, such as sliding motion. We present a modular framework for discovering such systems by first estimating switching hyperplanes from data and then learning smooth dynamics within each region using geometry-constrained neural networks. The geometry-learning phase is studied from a statistical perspective, analysing the identifiability of the discontinuities and the robustness of the procedure. We also introduce a novel neural network architecture with a prescribed discontinuity set, and provide a theoretical analysis of its approximation properties. The approach is tested on low-dimensional benchmark problems, including dry-friction oscillators and the PP04 climate model for the ice ages.

LLM agents have achieved strong performance on general software engineering tasks, yet struggle with domain-specific code generation. We identify the root cause as the agent's lack of tacit knowledge, including domain-specific business rules, interface contracts, and operational conventions that developers internalize through practice but never document. This knowledge is deeply buried beneath the domain code, dispersed across code entities and their dependency relations, and invisible to the agent that lacks it. These properties make tacit knowledge inherently difficult to retrieve or learn. In this work, we propose PRAXIS, a framework that enables agents to systematically extract, represent, and reuse tacit knowledge for domain code generation. PRAXIS acquires tacit knowledge by simulating human development workflows within the target codebase, distills it into structured units organized on the code dependency graph, and proactively surfaces it to the agent at the point of code interaction. Extensive experiments demonstrate that PRAXIS outperforms state-of-the-art agents equipped with powerful agentic search capabilities, as well as experience-based and skill-based methods. The approach integrates seamlessly into various agent frameworks and LLMs with consistent performance improvements, and supports continual evolution with performance steadily scaling as practice accumulates.

Existing optimal transport (OT) models primarily seek an OT map or plan between distributions by minimizing a prescribed transport cost or distortion. However, minimizing transport cost or distortion alone may fail to identify a geometrically meaningful transformation between the two distributions. To address this limitation, this paper proposes a novel coupled OT framework that leverages a small number of annotated landmarks to guide the recovery of an underlying deformation governing the distribution transformation. The coupled OT framework integrates the optimization of the transport plan and the deformation field into a unified model, where the landmark-guided deformation field and the cost-driven transport plan are coupled through a mutual-consistency constraint. As a result, the deformation is jointly determined by the annotated landmarks and cost-driven distribution matching. The proposed framework provides a principled connection between landmark-based registration and transport-based distribution matching, enabling the recovery of transport maps from sparse geometric supervision. We establish the well-definedness of the proposed model in a general variational setting and develop a finite-element-based numerical algorithm for computation whose convergence properties are systematically analyzed. The practical effectiveness of the proposed approach is verified in shape matching.

World models provide digital simulacra of the true world, allowing agents to be trained and tested before costly real-world deployment. At each time step, they receive an action and generate an observation and reward matching the statistics of the true world. In complex environments where present outcomes depend on events far in the past, this requires memory. One might expect that, by increasing memory, we can always build a model accurately enough to align the optimal agent policies of the real and virtual worlds. We show that this is false for classical world models, even when the true world itself is classical. We construct true worlds for which every finite classical model fails along the same possible trajectory: it either loses the ability to distinguish actions when the true world clearly prefers one, or repeatedly assigns the highest expected reward to suboptimal actions. Its expected-reward estimates also retain a nonvanishing average error. In contrast, each such true world admits a quantum world model using a single qutrit that reproduces it exactly: its reward estimates and preferred actions always match those of the true world, ensuring that the optimal policies of the real and virtual worlds remain perfectly aligned.

Pedestrian simulators need a behaviour rule for every agent, but privacy usually limits the data for setting one to aggregate statistics, namely zone-level device counts and origin-to-destination (OD) flows, with no individual trajectories. Such aggregates under-determine individual behaviour, because many different sets of decisions reproduce the same counts. We fine-tune a language model crowd agent so that the simulated population matches the observed destination composition, the fraction of the departing crowd heading to each point of interest. We read this target from the OD flow and reweight the model's own destination distribution onto it by iterative proportional fitting. Because fine-tuning inflates the dominant destination class, we fit the low-rank adapter to trajectories resampled to a corrected training composition that reaches the target after this inflation. On mobile network counts from two baseball games the fine-tuned agent runs without inference-time correction, cutting the destination-share error by 25%, while the grid correlation remains similar across policies.

The integration of inverter-based renewable energy sources into electric grids challenges conventional power system protection. Machine learning-based solutions can address these challenges by utilizing available data in modern smart grids. However, the lack of open datasets prevents reproducibility and fair comparisons between different approaches and their results, which hinders further progress. Therefore, this paper presents EvEMTBench, a synthetic dataset of faults and events generated using electromagnetic transient simulations. The physical plausibility of the power system simulation is ensured by validating the simulation parameters against established literature and providing comprehensive documentation of the simulation procedure. The dataset is designed for training, fine-tuning, and benchmarking machine learning models by providing synchronized point-on-wave voltage and current measurements at 9600 Hz across a diverse set of topologies and voltage levels. The inclusion of a wide range of fault and operating events allows the utilization of EvEMTBench for different tasks like incipient fault detection, fault localization, or event detection.

Current dexterous grasp planners primarily optimize for physical stability, focusing on whether an object can be grasped rather than how it should be grasped to support downstream functional tasks. However, conditioning grasp synthesis on specific human grasp taxonomies typically requires prohibitively expensive, object-annotated datasets. To address these limitations, we propose CoToGrasp, a novel generative framework that synthesizes diverse, stable grasps strictly conditioned on specific contact topologies. To bypass the data collection bottleneck, CoToGrasp is trained entirely in an object-agnostic manner. We introduce a feature-based canonical workspace that projects local object features into a unified gripper-centric domain, effectively decoupling the semantic functional intent from the arbitrary object geometry. By learning the intrinsic contact manifold of the gripper within this workspace, our model achieves zero-shot generalization to unseen objects at inference. Extensive evaluations on the large-scale DexGraspNet dataset demonstrate that CoToGrasp achieves state-of-the-art performance, outperforming existing taxonomy-guided planners. Finally, we demonstrate the physical viability and kinematic feasibility of our synthesized contact topologies on a physical robot platform. Code is available on our project website at https://cea-list.github.io/cotograspweb/ .

Fast and accurate segmentation of Acute Ischemic Stroke (AIS) lesions is essential for stroke prognosis and treatment planning. Non-contrast CT (NCCT), the first-line imaging modality for diagnosing ischemic infarcts, exhibits subtle infarct contrast, making manual delineation slow and labor-intensive. Motivated by this, and by the clinical practice of comparing brain hemispheres to localize infarcts, we propose a two-stage, nnU-Net-compatible 3D segmentation method. The first stage corrects head tilt to align each scan to its true anatomical mid-sagittal plane; the second applies a novel Asymmetric Feature Extraction (AsymFeX) module, comparing each voxel to its true contralateral counterpart within a local 3 x 3 x 3 neighborhood via cross-hemispheric attention, feature disparity estimation, and dual-scale gating to capture both large and small infarcts. On AISD, our method achieves 0.6796 Dice, 23.53 mm HD95, and 7.69 mL AVD, significantly outperforming existing state-of-the-art methods, with clinically relevant volumetric analysis at the 70 mL thrombolysis-eligibility threshold. Proof-of-concept evaluation on ATLAS v2.1 and ISLES'24 demonstrates that the same symmetry-driven design generalizes across imaging modalities and stroke time points without architectural changes, further supported by an uncertainty analysis assessing reliability under clinical deployment. Code is publicly available at https://github.com/biomedia-lab/AIS-detection.

Skchange is an open-source Python library for detecting structural changes in time series. It implements modern change detection algorithms within a unified and extensible framework. The algorithms are modular and composable, and they include changepoint search methods based on both cost minimisation and statistical tests. Key features include the detection of anomalous segments in addition to changepoints; theoretically well-founded fast and approximate search methods; theoretically well-founded algorithms for high-dimensional data, covering settings where either few or many features change simultaneously; utilities for automatic and data-driven penalty calibration, which balances false alarms against missed detections; and a large collection of built-in costs and statistical tests. The design follows established scikit-learn conventions to streamline both user and contributor experience, and Numba is used extensively to achieve high computational performance. Source code and documentation are available at https://github.com/NorskRegnesentral/skchange.

Self-supervised pretraining on remote sensing imagery typically treats all samples as equally informative, despite large variability in geographic and visual structure. We propose a curriculum learning strategy for self-supervised Earth observation that ranks samples by geographic isolation, a label-free proxy derived entirely from geolocation metadata already present in geospatial datasets, requiring no image decoding, no model feedback, and no manual annotation. Unlike visual complexity proxies, it scales as O(D log D) with dataset size D and is well-defined for both contrastive and reconstructive objectives. We integrate the proposed measure into MoCoV2 and MAE pretraining and evaluate across three downstream tasks from CopernicusBench (BigEarthNet, DFC-2020, LCZ). Our curriculum reaches baseline final-epoch performance using as few as 20% of the training budget (MAE) and at most 40% (MoCo) of the training budget, and improves final downstream performance by up to +5 mAP on BigEarthNet, with gains of 1-5 points across benchmarks, matching visual-complexity curricula while reducing pre-computation cost by more than 140x (4 s vs. 568 s on SSL4EO). A CKA and effective-rank analysis further reveals that curriculum-trained encoders develop higher-dimensional, more uniformly utilized embedding spaces throughout training.

Origami metamaterials offer significant potential for stiff deployable structures However, fabricating load-bearing cellular structures from thick panels introduces geometric interference at non-manifold junctions. Conventional thick-panel fabrication often disrupt ideal kinematics, thereby compromising smooth motion and scalability. This study proposes a modular fabrication framework that preserves one-degree-of-freedom rigid-folding kinematics in thick and non-manifold origami metamaterials. By decomposing non-manifold junctions into a hierarchy of stacked, modular hinged panels, our approach successfully accommodates synchronized hinge motions using scissor-like linkages. Exploiting this representation, we implement a graph-based topology optimization framework that tailors macroscopic stiffness while preserving folding connectivity. We demonstrate this approach by fabricating optimized prototypes that deploy seamlessly with a one-degree-of-freedom motion. Furthermore, we demonstrate engineering scalability through the large-scale construction of extensive deployable systems assembled from modular panels, which exhibit high load-bearing capacity. These results pave the way for the practical fabrication of structural, large-scale deployable metamaterials.

A minibatch can influence training beyond the update at which it is observed because AdamW stores past gradient information in its optimizer states. We study this delayed effect through paired trajectories that differ only in one gradient update and share the same subsequent training sequence. We formulate AdamW as a finite-horizon input–state–output (ISO) system whose state contains the model parameters and first- and second-moment estimates. Linearizing the joint dynamics yields a signed response operator that maps a localized gradient perturbation to its future loss effects, revealing how optimizer memory shapes their magnitude, timing, and sign. We further derive an exact multistep error decomposition and establish first-order finite-horizon accuracy under local smoothness and controlled activation switching. Experiments validate the response mechanism and optimizer-state effects, while repeated-future analyses reveal substantial prospective structure in delayed influence that can be partially recovered from ISO approximations. Code is available at https://github.com/Kanyooo/Loss_ISO.

Step-level credit evaluation can conflate numerical association with ordering and branch recovery with position-selection value. We test these distinctions in ALFWorld with two independently seeded replay banks: 27,520 continuations over 172 fixed turns. At 16 continuations per arm per bank, cross-bank replay-contrast associations are Pearson 0.736 and pooled Spearman 0.330. On 30 historical failures, A-count-selected branches at confidence-/implicit-selected positions gain 8.54 percentage points over same-position factual reruns, but their 3.33-point advantage over frozen-random positions under the same alternative-selection rule remains unresolved. Numerical association alone does not establish reliable pooled ordering; branch recovery alone does not establish better score-informed position selection. Historical judge alignment is modestly positive with an unresolved archived-control increment. Same-cohort C/D information-source comparisons and a short, sparse-exposure training pilot provide scope checks; incremental hindsight training utility remains unestablished. Findings are conditional on the tested policies, intervention support and finite budgets.

Multifingered grasping is a crucial robotic skill, but current deep-learning grasp planners often struggle to generalize to new objects because they are trained on limited, object-specific datasets. We introduce a fundamentally different approach, grounded in the observation that the gripper and the object share identical surface geometry at their mutual contact points. We propose GOAG: Generative and Object-Agnostic Grasp Planner for Dexterous Robotic Manipulation, a novel deep generative model that learns a compact latent representation of a specific gripper's contact surface distribution, enabling the efficient sampling of valid grasp configurations without relying on object-specific training data. We show that by introducing object features only at inference time, our model can effectively retrieve admissible contact areas that are compatible with the gripper's capabilities. We validate our approach through extensive experiments on established grasp protocols in both simulated and real-world scenarios, demonstrating its effectiveness with different grippers from the literature. Our method delivers state-of-the-art results on the objects from the MultiDex dataset, achieving an average success rate of 86.93%. It offers significantly faster processing when generating numerous grasps, while matching the performance of leading approaches specifically trained on this dataset. Unlike these methods, our approach does not rely on object-specific training data, highlighting the advantages of object-agnostic learning. It effectively addresses the generalization challenges faced by traditional data-driven grasp planners. Code and videos are available on our project website https://cea-list.github.io/goagweb/ .

Long-context modeling is a pivotal capability for Large Language Models, yet the quadratic complexity of attention remains a critical bottleneck, particularly during the compute-intensive prefilling phase. Our previous work, FlashPrefill, mitigates this cost through instantaneous pattern discovery and max-based dynamic thresholding; however, it remains an algorithmic prototype that is still distant from production deployment. In this paper, we present FlashPrefill V2, which evolves FlashPrefill from a prototype toward practical long-context serving along three dimensions. First, we introduce a mean correction term that effectively suppresses the approximation error, keeping performance degradation manageable even at extreme sparsity levels. Second, we redesign the sparse attention operator with PackGQA memory access, warp specialization, and pingpong pipelining, fully aligning with the latest FlashAttention-3/4 implementations and supporting FP8 inference to meet practical quantization requirements. Third, FlashPrefill V2 natively supports paged KV cache and continuous batching, allowing integration as an attention backend in modern inference frameworks such as SGLang. Extensive evaluations on NVIDIA H20 GPUs—among the most widely deployed inference accelerators—demonstrate that FlashPrefill V2 delivers up to 47.26x and 27.19x speedups over FlashAttention-2 at 128K context length under FP8 and BF16 precision, respectively, and, in FP8, still achieves a 30.49x speedup against an FA3/4-aligned dense baseline.

This paper proposes a controller design method for shaping the controllability Gramian into a desired form to design the effect from exogenous inputs to the system state. Using the Bures–Wasserstein distance, we formulate the shaping problem as the minimization of the distance between the system Gramian and a desired Gramian, and the objective function is shown to be strictly convex on the set of symmetric positive definite matrices. In addition, by deriving a semidefinite programming formulation via a linear matrix inequality (LMI), computational efficiency is improved and additional LMI constraints can be incorporated. When the exogenous input is modeled as Gaussian white noise, the proposed framework is closely related to $H_2$ control, which can be interpreted as a special case of optimal transport. Numerical examples demonstrate anisotropic controllability design for a guidance robot and verify the ability to impose additional directional constraints through LMIs. The numerical examples also confirm that the proposed method approaches $H_2$ control as the desired Gramian tends to zero.

Testing embedded software in modern vehicles is challenging due to system complexity, decentralized architectures, and strict safety and performance constraints. In this work, we present an end-to-end, deployment-aware testing pipeline for IoT-based automotive applications. The pipeline combines requirement-driven test and code generation with large language model (LLM) and vision-language model (VLM) assistance, and human-in-the-loop curation to reduce manual effort and improve consistency. Using Eclipse openDuT, it supports flexible, distributed deployment across geographically separated cyber-physical and IoT infrastructures, optimizing for node availability and cross-organizational coordination. For validation, we conduct a case study using a Child Presence Detection System (CPDS), achieving full functional requirement coverage across all 9 requirements and 100% Gherkin generation accuracy on the controlled requirement set. Distributed test execution across geographically separated ECUs via Eclipse openDuT confirms the pipeline's applicability to OEM–supplier testing workflows.

Micro-View Order-Dispatching assigns available drivers to passenger orders within each dispatch batch and is critical to the service quality and operational efficiency of ride-hailing platforms. Mainstream industrial solutions follow a multi-stage paradigm of model prediction, value calculation, and dispatch matching. Although dispatch quality is determined by the final batch-level assignment, these stages optimize different intermediate objectives. This cross-stage objective inconsistency means that improving a single stage does not necessarily improve the overall dispatch result. We therefore formulate Micro-View Order-Dispatching as a generative matching problem and propose GenMatch, an end-to-end Generative Matching framework and the first such framework deployed in a real-world production environment. Applying generative modeling to this problem introduces three challenges. First, each dispatch batch forms a dynamic sparse bipartite graph, requiring efficient structured batch-level encoding. Second, replacing the hand-crafted value function requires learning unified business utility from heterogeneous feedback. Third, directly generating an assignment requires tracking the evolving matching state because each selected order-driver pair changes the remaining feasible candidates. GenMatch addresses these challenges with a Context-Aware Bipartite Encoder, a Business-Aware Utility Learner, and a State-Aware Pointer Decoder. Extensive offline evaluations and online A/B tests in five cities across DiDi's international ride-hailing markets show consistent improvements over competitive baselines, confirming the effectiveness and practicality of GenMatch for industrial order-dispatching.

Advanced Persistent Threat (APT) attacks pose a critical challenge to modern systems, as their stealthy, multi-stage nature renders conventional detection methods ineffective. While provenance graphs provide rich behavioral context for attack investigation, attack-relevant evidence is often sparse and embedded in large volumes of routine system activity, making full-graph learning both computationally expensive and difficult to correlate over long attack sequences. We present TGL-APT, an adaptive investigation framework built on the observation that attack-relevant information is non-uniformly distributed and often mediated by structurally influential or behaviorally distinctive entities, which we characterize as information-bottleneck nodes. TGL-APT combines three complementary components: (1) information-bottleneck-guided graph distillation that suppresses provenance redundancy while bounding structural distortion and preserving causal reachability; (2) adaptive temporal graph learning that continuously refines the core node set as node relevance evolves; and (3) cross-spatiotemporal attack fingerprint alignment that associates fragmented suspicious activities across different entities and time windows. Finally, causal expansion and stage characterization reconstruct coherent attack processes for investigation. Experiments on three DARPA E3 datasets show F1-scores of 95.7%, 90.9%, and 88.9%, while reducing training time, detection latency, and memory usage by approximately 39%, 33%, and 22%, respectively, compared with KAIROS. These results demonstrate that TGL-APT effectively balances detection performance, computational efficiency, and investigation capability for provenance-based APT analysis.

Inference-time selection methods, such as Best-of-N, improve generation by sampling a pool of candidates and selecting the top-ranked completion according to a reward model. Distillation seeks to amortize this procedure into a single policy by replacing raw rewards with in-pool ranks and learning a policy that upweights higher-ranked completions. However, existing rank-based policies typically use smooth full-support reweighting, so low-ranked completions receive less mass but remain in the target support. Although a sharper reweighting reduces lower-tail mass, it also increases reliance on brittle ranking at the top made by a single reward model. We propose TUP: a Truncate-bad, Upweight-good Policy that removes low-ranked completions from the support and reweights only the retained upper tail with a tunable sharpness. TUP admits a closed-form, prompt-independent normalization and can be trained fully offline via binary cross-entropy, using shifted-truncated win-rates as soft labels and distilled-to-reference log-likelihood ratios as logits. Theoretically, under certain assumptions, we show that for any unknown oracle reward, the best monotone rank-reweighting can be matched by a lower-tail truncation rule, providing formal support for removing the lower tail rather than merely downweighting it. Empirically, we show that TUP is competitive with strong offline alignment baselines.

Personalized text generation aims to make LLMs write in a specific individual's style, yet existing benchmarks measure task accuracy or preference alignment rather than whether the model's output actually resembles the target author's writing. We introduce PersonalBench, a benchmark that evaluates inference-time personalization methods through three independent lenses: LUAR (a trained authorship verification model), an LLM-as-judge, and automated stylometrics. Across 50 authors, 1,000 generations, and two model families (Qwen 3, GLM-4), we find that personalization methods do produce author-differentiated output (LUAR discriminates target authors within generated text at AUC=0.918) but this differentiation never crosses the human-LLM boundary. All methods achieve LUAR similarity to real authors in the range 0.484-0.508, below the cross-author human floor of 0.626 (ceiling 0.756). The LLM's own authorship fingerprint dominates: generated text is more distant from any human author than random humans are from each other. Methods are statistically indistinguishable from each other on LUAR (spread 0.024) despite appearing differentiated on the LLM judge, a discrepancy we trace to circularity between trait extraction and profile extraction. We validate that LUAR reliably measures authorship in our corpus (AUC=0.76 single-post, 0.96 multi-post). We release PersonalBench as a calibrated measuring stick: inference-time personalization modulates the LLM's style but does not bridge the gap to human authorship.

We describe our entry to the PhysAI Dynamic 4D Reconstruction Challenge, which placed third of 27 teams at 0.58356 APD on the final leaderboard, without a single gradient update. This was not the plan: of thirteen fine-tuning configurations of a pre-trained 4D backbone, twelve degraded the challenge score, and eleven of those twelve improved local validation at the same time. We trace this inversion to the structure of the benchmark: only 25% of the evaluation set belongs to the data variant released for training, so updates that fit the available data damage the pre-trained features the remaining 75% relies on. Our system therefore freezes the backbone and spends its budget at inference time, fusing three decoding configurations – temporal stride-3, horizontal-flip test-time augmentation, and dense stride-1 – under a convex weighting. The ensemble recovers +0.041 APD over the frozen baseline, more than any training run achieved, at zero training cost.

Recent agent benchmarks increasingly ground evaluation in executable environments, from code repair to web navigation, app APIs, and function calling. Yet completing consequential work beyond code requires more than producing a plausible response or valid tool call: agents must gather missing information over multiple turns, follow domain policies, coordinate dependent tools, and realize the correct persistent state transition without collateral effects. In this paper, we introduce Thinkingbox, a sandbox for tool-agent-user interaction that provides isolated MCP-compatible tool sessions, complete execution traces, and outcome evaluation over terminal backend state. Built on this sandbox, Thinkingbox-bench contains 507 policy-conditioned workflows across business scenarios, including retail, hospitality, auto insurance, neobank internal IT, and consulting IT/HR support. Each attempt is evaluated by task-specific executable checks that accept valid trajectories while rejecting wrong, missing, or extra effects; designated tasks additionally check required properties of the final response. Our experiments reveal that even the strongest proprietary and open-weight models show steep reliability drops: Claude Opus 5 falls from 66.50% pass@1 to 47.53% pass^20, and Kimi-K3 from 57.37% pass@1 to 17.60% pass^20. Moreover, many failed trials terminate cleanly after valid state-changing actions, so response- or tool-call-level signals poorly proxy end-to-end completion. Thinkingbox-bench reveals a large gap between occasionally finding a successful trajectory and reliably completing stateful business tasks. We release Thinkingbox, Thinkingbox-Bench, and Thinkingbox-integrated reinforcement learning framework.

This paper presents an open-source software stack that restores ROS 2 support for the Franka Emika Panda robot while resolving the long-standing unreliability of its external position control interface. We first analyze the root causes of unstable position control and show that the observed vibrations and protective stops arise from the timing of the external control loop and sampling jitter, rather than from limitations of the robot itself. Building on this analysis, we introduce an asynchronous hardware interface that decouples real-time communication from the ROS 2 control loop, a rate-matching mechanism for slower command sources, and a position-domain reference generation strategy that produces reliable, smooth position commands. Experimental validation shows that the proposed architecture reliably tracks velocity references by reducing motion artifacts introduced by the official implementation, and the stack is validated across motion planning, compliance control, position-controlled manipulation, and haptic teleoperation on two independent Panda platforms. By restoring a modern, reliable, and open ROS 2 ecosystem for the Panda, this work lowers the barrier to developing safe, responsive, and reproducible human-robot collaboration applications that integrate planning, perception, interaction, and shared autonomy. Code and videos are available on our website at https://sites.google.com/unisa.it/fer-ros2.

Multimodal LLMs can see a document, but they often can't read it reliably. Small text, tables, visual cues, and topological elements still trip them up under direct visual inference, even when the page is already sitting in the model's context. Most document-VQA systems treat perception as fixed: they encode the page once, ask the question, and answer from whatever the model happened to extract in that single fast pass. We think document VQA needs slower, more deliberate perception: rather than answering from one fixed encoding, the model should spend a bit of extra compute at inference time working out what to look at next, and only then answer. We build this into Q-Guide, a small agent that reads a question, works out what evidence it is still missing, and calls targeted tool(s) to recover it—reading text where text is needed, zooming in where detail is needed, or grounding a region where position matters. On DocVQA2026 and Manga109, Q-Guide outperforms both direct prompting and recent multi-agent document systems ($65.0\%$ vs.\ $40.0\%$ on DocVQA2026, $32.4\%$ vs.\ $24.4\%$ on Manga109), and the improvement holds across three Claude backbones (Opus 4.6, Sonnet 4.6, and Opus 4.5). We find that accuracy scales with the perception budget—most of the gain appears within two to three deliberate rounds—and that the gain comes from directing perception to the right place, not from complex control logic: adding planners, routers, or multiple collaborating agents does not help.

Full-cycle biventricular geometry is essential for characterizing cardiac function. However, dense and temporally consistent 3D+t biventricular meshes are not routinely available, whereas end-diastolic (ED) anatomy can often be obtained reliably. We therefore investigate full-cycle biventricular motion synthesis from a single ED mesh. This task is challenging because cardiac deformation is spatially heterogeneous and phenotype dependent, while conventional global generative models often obscure localized motion patterns. In this study, we propose a region-specific and phenotype-adaptive framework that integrates motion-informed functional parcellation with conditional latent flow. A functional partition learned from reconstructed motion organizes the ventricular surface into regions with coherent dynamics and enables topology-aware regional feature exchange. A phenotype-conditioned rectified-flow model subsequently maps the ED anatomy to full-cycle motion latents through fine-grained conditioning and prototype-routed motion adapters. An optional control branch further incorporates available motion descriptors for controllable synthesis. Experiments on ACDC, M&Ms, and M&Ms-2 demonstrate consistent improvements in geometric accuracy and functional fidelity. Under ED-only synthesis, our method achieves biventricular ASSD, HD95, and vRMSE of \(1.49\pm0.34\)~mm, \(3.77\pm1.06\)~mm, and \(3.31\pm1.03\)~mm, respectively, outperforming all competing methods. Complementary functional and robustness evaluations further demonstrate that the synthesized sequences preserve physiologically plausible ventricular dynamics and generalize across cohorts and disease phenotypes. The code will be released publicly upon acceptance of the manuscript for publication.

Large vision-language models (LVLMs) have achieved remarkable progress in video understanding and reasoning. Despite extensive studies on text- and image-based jailbreaks, video jailbreaks against LVLMs remain largely unexplored. Existing video jailbreak methods mainly manipulate textual content embedded in videos, while overlooking how such information is organized over time. Our analysis reveals that jailbreak effectiveness depends not only on the semantics of textual information but also on its temporal presentation, including duration and timing-slot allocation. Motivated by this finding, we use subtitles, which are common in real-world videos and allow semantic content to be presented under precise temporal control without appearing visually intrusive, as a natural attack medium. Based on this insight, we propose TempJail, a black-box video-based jailbreak framework that constructs query-aligned dialogue-style subtitle sequences and optimizes their temporal scheduling to exploit temporal vulnerabilities in LVLMs and elicit responses that satisfy the harmful intent of the source query. Extensive experiments on four representative LVLMs and two datasets demonstrate that TempJail achieves the highest attack success rate across all evaluated model–dataset settings, outperforming the strongest baseline by 53 and 18 percentage points in dataset-averaged ASR on GPT-5 and Gemini 3.5-Flash, respectively.