Papers

A useful model of AI inference infrastructure must specify the system state, the information available to an observer, and the decisions the model is intended to support. We introduce AID (AI Infrastructure Dynamics), a framework for describing this learning problem across coupled physical, computational, networking, and serving processes. The formulation allows structured and variable-size state, asynchronous observations, multiple physical timescales, and demand that responds to service. We distinguish representations that support prediction under an existing policy from those that preserve service outcomes under changed actions, and separate both from identifying intervention responses. Two analytical results describe a lower bound on prediction error when available observations cannot distinguish models and a sufficient condition for exact controlled state reduction. These results apply established information and state-abstraction principles to AI infrastructure. We then describe a validation protocol for cache representations, workload histories, measurement availability, and imposed actions.

Graphical consoles offer a practical interface for medical acquisition assistance, allowing agents to work through the controls and visual feedback used by human operators. Reliable assistance requires linking on-screen anatomy to acquisition decisions that determine what image evidence becomes available next. We introduce MedImageOSWorld, a benchmark for evaluating this capability in simulated CT, MR, and ultrasound consoles. Using screenshots and mouse-and-keyboard actions, agents configure protocols, plan acquisitions, inspect the resulting images, and make corrective adjustments across seven capability levels, from console operation to feedback-driven control. Evaluation combines task-specific workflow checks with hidden anatomical ground truth to assess procedural completion and acquisition outcomes separately. A common evaluation protocol specifies episode conditions and interaction budgets, while recorded trajectories support analysis of how agents observe, act, and respond to acquisition feedback. Across eleven open-weight agents, success rates range from 3.0 to 25.0 on a 0-100 scale while workflow-progress rates reach 21.5-74.5: agents complete much of the console workflow but rarely acquire the intended anatomy. Success collapses between perception and planning, from 79-90% at the lowest three levels for the best agent to at most 12% for millimetre-level planning and 0% for closed-loop control among open-weight agents. Two proprietary agents reach success rates of about 34 and exceed the best open-weight agent mainly in acquisition quality (61 versus 42). MedImageOSWorld provides a controlled setting for studying whether general-purpose GUI agents can translate visual observations into effective medical acquisition decisions.

Predicting security incidents is a profound task critical for informing proactive defensive measures and cyber-insurance policies. Prior work tackling this problem mainly utilized structured, manually defined features based on network measurements (e.g., protocol misconfigurations). Still, despite leading to promising performance, the network-based features may fail to capture aspects related to adversaries' motives. To fill this gap, our work leverages geopolitical data mined from public sources–which may help capture attacker motives–to forecast security incidents. Specifically, our approach relies on news articles and transcribed podcasts that are fed to large language models to automatically produce rich representations. The representations are then fed to a classifier trained to forecast future incidents based on historical ones. Our evaluation with a large incidents dataset (>15,700 records) demonstrates substantial accuracy (71.4% ROC AUC) with geopolitical data alone. Notably, combining geopolitical data and network measurements outperforms the state-of-the-art technique based on network features alone (81.3% vs.77.5% ROC AUC). Our analysis also helps shed light on when geopolitical data is most helpful and the data sources that are most useful for accurate forecasting.

Event-informed forecasting requires translating reports and historical responses into changes to a numerical forecast. We propose TIMBRE (Temporal Integration of Memory-Based Responses and Evidence), which combines source-aware representation, state-conditioned response transfer, and reliability-guided fusion before a frozen forecast head. A separate readout adjusts interval widths while preserving the median. In a single-seed, one-epoch development study of 13 tasks, TIMBRE improves MAE over ordinary fusion on eight tasks but over native Chronos-2 on only two. Disabling response transfer in the trained model reduces BTC and AULF MAE by 47.04% and 6.81%, respectively. These findings identify sensitivity to learned response transfer rather than a general forecasting advantage. Missing development-set scores and the absence of retrained ablations limit attribution to individual evidence mechanisms.

An agent with long-term memory can answer from a record it should no longer use, such as a plan the user later cancelled. We ask what a memory store must expose for an agent to know this before retrieving anything, and we score that judgment on its own, as metamemory monitoring. Readers see only a value-free summary: record counts by lifecycle state and a list of attribute names. Each of 148 questions is asked against three versions of one store that differ in one attribute, so wording cannot give the answer away. What a model adds depends on the length of that list. On the short lists the benchmark's ground truth produces (2.6 names on average), none of five language models reliably beats a cosine-similarity lookup over the names, and the three stronger ones are equivalent to it within 0.05. Under a control that fixes counts and list length, none is reliably above it and none is shown equivalent. On lists longer than the stores we tested write, padded to 60 names, the lookup loses 0.18; GPT-5.6 Luna and Sol lose 0.06 to 0.07 and lead it by 0.13 to 0.14, while the other three fall with it. The lead holds for Sol against a sibling padded to the same length and counts. Reworded to share no word with the names, the lookup loses 0.02 and one stronger reader edges 0.04 ahead. On the short lists the three stronger readers lead only where the summary counts a state without naming it, and one count feature closes that lead within a question, though not when items are pooled. The backends we tested omit or misstate this information: under FR-Bank's own metadata, GPT-4.1 mini, Haiku 4.5 and the lookup fall from 0.73, 0.82 and 0.75 to 0.60, 0.71 and 0.59. In these tests, what the store wrote down limited every reader we gave it to. All stores are synthetic; control, whether a better judgment yields a better answer, is left to a later study.

Recent incidents show that AI agents sometimes reach measured goals through unsanctioned means. This study applies self-control, general strain, anomie, neutralization and routine activity theory to reward hacking in generative AI models, and it treats the measures as behavioral analogues. Study 1 (2,310 conversations, seven models) measured delay discounting with the Kirby Monetary Choice Questionnaire and stated willingness to take shortcuts. Pressure raised the discount rate k 2.8-fold in fresh conversations but 12.6-fold when the same sentence followed a baseline answer, which indicates a response to conversational cues rather than a stable trait. Models chose a shortcut in 1 of 700 dilemmas when answering as themselves and in 64 of 700 when asked to assume human impulses, each step of pressure raised the odds by 40%, and shortcut answers contained far more techniques of neutralization (rate ratio = 146). In the preregistered Study 2, five models worked on 20 coding tasks whose tests contradicted their specifications. Two Claude models never cheated. GPT-5.6, Qwen and DeepSeek cheated in 86%, 69% and 65% of episodes and clearly disclosed the conflict in 27%, although their reasoning recognized it in 95%. GPT-5.6 had never endorsed a shortcut in Study 1. The registered effects of pressure and of an auditor cue did not survive correction for multiple testing. In exploratory analyses, two further models cheated in 69% and 100% of episodes, and one sentence stating that the specification takes priority eliminated cheating in all 280 episodes. Therefore, stated refusal does not guarantee compliant agent behavior.

Missing modality remains a longstanding challenge in multimodal learning. Existing methods typically address this issue through modality recovery or adaptive strategies. However, they overlook models' internal cross-modal dependencies formed during multimodal training, which later impair robustness. We systematically characterize a counterintuitive deployment-time failure mode: models trained on full modalities can underperform unimodal models when one modality is missing at inference time. This pattern appears across diverse architectures, such as fusion models, CLIP-style two-tower models, and vision-language models. We show that such degradation is closely associated with learned cross-modal dependencies in the principal parameter subspaces. Multimodal training induces structured rotations of these subspaces, particularly in cross-modal interaction layers. These rotations are associated with reduced task-aligned margins and larger task-aware representation harm under missing-modality inputs. We propose Geodesic Unlearning (GU), a lightweight parameter-editing method that leverages Grassmannian subspace geometry for structured subspace correction to improve missing-modality robustness. It rotates the principal input subspace toward a unimodal reference along a geodesic path. We prove that this correction minimizes the distance to the reference within a fixed subspace-distance budget. Experiments across architectures and datasets show that GU improves performance under missing-modality inference while preserving full-modality accuracy, outperforming strong missing-modality robustness baselines. These findings support a geometric view of deployment-time missing-modality degradation and suggest localized subspace editing as a practical route for robustness correction.

At the core of modern prompting techniques is contextual sensitivity, the ability of large language models to adapt their predictions based on inference-time context. Despite its central role, inference behavior under strong contextual influence remains poorly understood, particularly at the level of internal inference dynamics. We introduce a theoretical framework for analyzing contextual influence through inference dynamics, enabling quantitative characterization of inference behavior beyond output-level answer changes. Our analysis shows that inference dynamics do not exhibit unbounded drift under repeated contextual assertions. Instead, predictive representations converge to stable, query-dependent regimes that fundamentally constrain whether contextual signals can alter a model's prediction. This leads to a surprising finding: Repeated contextual assertions do not act as accumulating evidence during inference and may therefore fail to alter a model's prediction even under unbounded repetition, while in other cases a prediction change becomes inevitable. We empirically validate our theoretical predictions, demonstrating strong alignment between theory and observed inference behavior. These contributions offer a principled pathway toward characterizing the limits of contextual influence during inference, providing practical implications for model development.

The adjoint method is a standard approach for computing gradients in PDE-constrained optimization and central to parameter estimation, model calibration, and optimal control. This tutorial presents a discrete, implementation-oriented introduction to the method with an emphasis on conceptual understanding. Starting from an inverse problem governed by a partial differential equation, it develops the corresponding optimization problem and derives the adjoint gradient from the perspective of iterative descent methods. The roles of the state, the parameters, the objective, and the constraint are made explicit, and the adjoint equation is obtained directly from the gradient computation rather than postulated. Particular attention is given to explicit derivations, implementation details, and the connection between the mathematical development and the resulting computational algorithm. The tutorial also examines the influence of discretization on the adjoint construction, explains the origin of the backward-in-time adjoint recursion, and presents practical procedures for gradient verification. A vibrating-string inverse problem serves as a running example, illustrating the complete workflow from model formulation to adjoint-based optimization. The presentation is intended for advanced undergraduate students, graduate students, and practitioners seeking a step-by-step understanding of adjoint gradient computation.

Neural Architecture Search (NAS) has proved to be a strong alternative to manual network design, but applying it to 3D medical image segmentation is limited by two well-known costs, large annotation budgets and multi-GPU clusters. Thus, this paper introduces Active-DiNTS (Active Differentiable Network Topology Search), an approach that embeds pool-based Active Learning (AL) into a bi-level differentiable topology search to perform architecture discovery and label curation jointly. At each query round, unlabeled MRI volumes are ranked by one of three uncertainty signals (Entropy, Variance, or Standard Deviation), and only the top-ranked volumes are sent to an oracle for annotation. The new labels feed two interlocked stages. Network weights are updated in an outer loop, while the macro/micro topology of a U-Net-style backbone is refined in an inner loop. Three AL regimes (weights-only, topology-only, joint) expose the speed-accuracy trade-off. Evaluations on the Medical Segmentation Decathlon (MSD) Task01 BrainTumour benchmark showed that Active-DiNTS surpasses DiNTS, C2FNAS, and nnU-Net in Dice, with gains of about 10 percentage points on Edema and 5 points on Non-Enhancing core, on a single GPU and using a fraction of the labeled volumes. The discovered architectures are denser and more FLOP-heavy than prior baselines, but remain competitive in trainable parameters and peak memory; the fastest search variant finishes in under 0.25 GPU-days, over 27x faster than the eight-GPU DiNTS search. Together, these results indicate that pairing differentiable NAS with active data acquisition is a practical recipe for accurate 3D segmentation under realistic constraints.

Time-series foundation models buy zero-shot forecasting with large temporal corpora; a vision model needs none, since a natural image implicitly embeds the patterns a forecaster must model, and an ImageNet-pretrained masked autoencoder forecasts a series by inpainting a rendering of it. A rendered series is not a natural image, however, and closing that gap takes temporal-aware adaptation. We show that the rendering geometry - how the series is folded and drawn - is a controllable, mixable axis for it. Kaleido detects the dominant periods, renders a rule-generated set of fold geometries, combines the inpaintings with a convex per-position gate fit on validation only, and fuses the result with the zero-shot output at one fixed share, with no per-dataset hyperparameter beyond the baseline's published settings. Training only LayerNorm (0.05%), Kaleido lowers MSE by 13% against the published zero-shot baseline on LTSF and, frozen, by 6.6%; on GIFT-Eval it improves the baseline by 7.4% in MASE and 19.3% in CRPS.

Camera-to-motion-capture (mocap) calibration is essential for using mocap as ground truth in robotics, AR/VR, and other computer vision tasks. However, the calibration can drift after deployment, while calibration residuals and visual inspection provide limited independent verification. We present Lollypop, a fiducial-mocap reference target for independent calibration verification. The target couples an ArUco fiducial with a mocap marker constellation so the visual center and tracked centroid represent the same physical point. Given a candidate calibration, verification projects the mocap point into the image and measures its disagreement with the detected fiducial center. Experiments show sub-pixel nominal error, sensitivity to controlled extrinsic perturbations, and increasing error during an illustrative mixed-handling sequence.

Traffic classification (TC) is crucial to secure Internet of Things (IoT) networks, whose edge nodes often operate under privacy, bandwidth, and energy constraints. Yet, encrypted payloads and limited computing power make accurate, real-time TC a challenging task. Existing learning-based TC approaches often fix the input configuration a priori, even though it directly influences both predictive performance and computational cost. This paper presents RepTC, a representation-aware hardware-constrained strategy that addresses session-level TC through joint model and input optimization. The method co-optimizes network architecture, session length, and header preprocessing; this enables joint control of model complexity, input scale, and data representation within a unified resource-constrained design space. The proposed approach enforces microcontroller-class constraints on memory, model size, and computation, and yields compact models deployable on low-power edge devices. A gateway monitors traffic by aggregating sessions and either performs inference locally or offloads to a low-power edge node. Both scenarios are validated on heterogeneous embedded hardware, including a Raspberry Pi 3B+, STM32 Nucleo-F401RE, and XIAO ESP32-C3, spanning Cortex-A, Cortex-M, and RISC-V processor architectures; measured inference latency ranges from 0.63 to 18.59 ms, with MCU-side inference energy between 0.50 and 1.41 mJ per session. RepTC yields configurations that achieve high accuracy on a variety of established benchmarks: 96.21% on ISCX VPN-nonVPN, 99.62% on USTC-TFC2016, and 99.97% on Edge-IIoTset; at the same time, model size and computational requirements were reduced by up to three orders of magnitude compared with state-of-the-art methods. The results show that representation-aware optimization can improve efficiency while preserving competitive classification performance.

Counterfactual explanations are most useful when they can be generated with low latency, remain close to the factual input, and satisfy input-domain, categorical, and actionability constraints. Achieving these objectives simultaneously is challenging for deep neural networks. Heuristic methods are often fast but may return invalid counterfactuals, whereas exact methods can certify global proximity but may not finish within practical time limits. We introduce DASH, a batched heuristic search method for finding close, valid, and actionable counterfactuals for deep neural networks under $\ell_1$, $\ell_2$, and $\ell_\infty$ objectives. DASH uses directional Lipschitz bounds and local affine models to generate anchors, then ranks and expands promising regions with batched network evaluations. We compare DASH against nine prior heuristic methods and time-limited exact mixed-integer baselines on four tabular datasets, with network depths from 2 to 32, and evaluate scalability on PBMC3k. Across 9,000 tabular query-norm cases, DASH returns a valid counterfactual within $5\%$ of the best heuristic-observed valid distance in $94.6\%$ of cases, with a median CPU search runtime of $0.061$ s. PGD-bisect, the baseline with the highest pooled within-$5\%$ coverage, meets this criterion in $41.1\%$ of cases, with a median runtime of $0.298$ s. These results show that the proposed search maintains high valid proximity across norms while keeping its search runtime practical.

In an image latent space, the embeddings of high-resolution, natural, and sharp images form a manifold. Degradation of high-resolution images pushes their embeddings off this manifold. Real-world super-resolution (SR) then becomes the task of mapping the degraded embedding back onto this manifold — not anywhere on the manifold, but to the point that preserves what the input still carries, both its semantics and pixel details. Every published method implements this mapping in a reconstruction-oriented latent space or pixel space. We claim these spaces are the wrong substrates for SR. Low-resolution and degraded images are embedded far from the manifold, making the mapping difficult and expensive. The lack of semantic information in these substrates also makes it difficult to navigate to the faithful point on the manifold, causing severe hallucination when degradation is heavy. Thus, restoring in a suitable latent space is crucial to the SR task. We show that the latent space of 23 fused layers of a frozen DINOv3-L is one such space that makes the SR task easier. Degraded images are embedded near the manifold. Moreover, this substrate contains a hierarchy of information, from pixel record to degradation robust semantics, guiding the model to find the faithful point on the manifold. On this substrate, a 415M decoder is trained under reconstruction and adversarial objectives to map the degraded embeddings back and decode to pixel space in one pass. The resulting model, RAESR, attains the best fidelity–perception trade-off among state-of-the-art adversarial and diffusion-based restorers on RealSR, DRealSR, LSDIR and DIV2K-Val, at 37 ms per 512 by 512 image on a single H20 GPU. Swapping the substrate for a VAE latent under an identical recipe loses on every metric.

Concurrent AI coding agents create a coordination problem in which cheap signals may prioritize work, but only an executable checker can establish the property being claimed. We study this separation through a property-scoped verification contract and MERGEGYM, a three-track benchmark for open-time scope forecasting, replay- conditioned conflict resolution, and scheduling. On a stratified 715-pair lineage set (167 textual conflicts), 79 conflicts (47.3%) occur despite disjoint authored file sets. This is an operationally important proxy mismatch expected from three-way merge: authored PR diffs are measured against PR-specific bases, whereas the checker compares both heads to their common merge base. A standalone lineage-union rule reaches held-out AUROC 0.877; a 12-feature logistic model reaches 0.882 [0.845, 0.917] with PR-AUC 0.661, and an untuned random forest on the same decision-time features reaches 0.902 [0.875, 0.928] with PR-AUC 0.701. At a 33.3% held-out replay budget, the logistic and random-forest models recover 81.2% and 82.9% of conflicts, respectively. Patch reconstruction succeeds for 48/79 zero- overlap conflicts and all 48 become clean; the other 31 cases are inconclusive, so this check validates the expected three-way-merge explanation rather than claiming a new Git mechanism. In T1, a zero-shot LLM reaches AUC 0.704 and LLM-plus- metadata fusion 0.740. In T3, a decision-time gate de-overlaps a median 91.7% of labeled scope collisions at 65.0% makespan inflation under frozen-label replay. Because local git merge-tree replay is already cheap in our logs (median 0.02 s), we do not claim that lineage triage saves this checker alone: when exact replay is cheap, verify everything. All empirical guarantees in this paper remain limited to textual mergeability or the explicitly stated frozen-label scheduling target.

Patient world models are increasingly proposed for longitudinal prediction, intervention-aware reasoning, and clinical-trial simulation. Causal or clinical intervention validity is distinct from predictive generalization and reliability; before making stronger claims, the underlying predictive state should generalize across patients, survive realistic shifts and missing observations, and expose failure through meaningful reliability signals. We evaluate these prerequisites in a deliberately narrow setting: short-horizon digital-biomarker forecasting from PhysioNet GaitPDB, comprising 165 participants, 306 recordings, and 51,129 context-future pairs. Using persistence, ridge, MLP, GRU, Transformer, and a compact JEPA-style predictor, we build an evaluation ladder that progressively removes raw temporal overlap, same-recording familiarity, and same-patient familiarity before testing unseen-patient generalization. For GRU, NMSE rises from 0.1227 under random-window splitting to 0.1393 after eliminating raw train-test overlap and to 0.1961 under patient holdout. Among 54 participants with repeated recordings, exposure to a different recording from the same patient improves GRU NMSE from 0.2177 to 0.1556, while a recording-excluded identity hypothesis is not supported at the participant level. Under participant-held-out evaluation, MLP and Transformer are statistically indistinguishable. Study shift, a four-times-longer prediction gap, and partial observation further degrade performance; under 50% temporal masking, Transformer NMSE rises to 0.611 while MC-dropout predictive variance falls. We do not claim a longitudinal or intervention-aware simulator. Instead, the results support a prerequisite evaluation stack of patient separation, repeated-measure controls, shift, missingness, and uncertainty validation before stronger patient-world-model claims are trusted.

Inflammatory bowel disease (IBD) requires frequent imaging-based assessment, yet interpretation of modalities such as wireless capsule endoscopy (WCE) and intestinal ultrasound remains heavily dependent on specialist expertise. Vision-Language Models (VLMs) demonstrate significant potential in multimodal medical image analysis, but their clinical adoption is hindered by their insufficient domain-specific reasoning, susceptibility to hallucination, scarcity of high quality training data in fine-grained diagnostics and limited interpretability. We introduce ARISE (Adaptive Agentic Reasoning with Image-grounded Self-Evaluation), an autonomous planning framework that models few-shot medical image understanding as a sequential agentic workflow. ARISE structures agent execution into a transparent 5-stage workflow: hypothesis generation, image-grounded evidence summarization, evidence-conditioned refinement, symbolic verification, and final diagnosis. We apply ARISE to IBD assessment across two independent patient cohorts: wireless capsule endoscopy (WCE) images for Crohn's disease and B-mode ultrasound data for ulcerative colitis. ARISE consistently improves diagnostic performance over baseline VLMs while exposing where reasoning succeeds or fails, providing a more interpretable basis for clinical decision support and realistic deployment.

This paper studies distortion-aware clustering for uplink fronthaul-limited cell-free MIMO networks employing maximum ratio combining (MRC). While MRC is appealing for its low-complexity, its performance is limited by interference, fronthaul distortions, and channel imperfections. This motivates clustering strategies that compensate for these impairments through coordination of access points. Since instantaneous small-scale fading-based optimization is not scalable in large systems, and is impractical due to frequent channel variations, we instead optimize an objective depending only on large-scale fading coefficients. To this end, asymptotic analysis is used to derive deterministic equivalent expressions for the average network sum rate, with focus on quantization distortion, leading to a quadratic-over-linear objective. Although maximizing such an objective under binary constraints is non-convex, we exploit its structure to develop a polynomial time scheme that attains the global optimum. Numerical results show that the proposed clustering method provides performance gains over improved variants of existing literature, and remains competitive with the small-scale fading-based global optimum obtained via exhaustive search.

In this paper, we present a distributed cascade force control system (DCFC) for multiple robots with the aim of pushing a rigid object towards a desired moving target without their inter-robot communication. These mobile robots are equipped with 360-degree vision-based soft tactile sensors utilized to determine contact location and resultant impact force. By investigating the dynamics of moving rigid objects on the flat, we proposed a distributed cascade control. The inner loop control incorporates contact force and positioning, ensuring the robots' pushing contact and application of the desired force to the object. The outer loop control coordinates the robots to push the object in a desired direction without inter-robot communication, regardless of the unknown object mass. The stability and convergence of the system are verified using the Lyapunov stability theory. We also conducted simulation and real-word experiments to validate the performance of the proposed control method, and the experimental results showcase the successful coordination of multiple robots in pushing an object towards a moving desired direction.

Analog computing with microwave signals can perform linear transformations directly in the analog domain, as the signals propagate through a microwave network. A fundamental question is which transformations can be computed with a given set of microwave components. In our previous work, we answered this question for networks of hybrid couplers and phase shifters by deriving a necessary and sufficient condition on the transformations these networks can compute, and we showed that the discrete Fourier transform (DFT) satisfies it. In this paper, we take a first step toward the formalization of analog computing with microwaves in Lean, a programming language and proof assistant that is increasingly adopted in mathematics. We formalize the considered components, their series and parallel connections, and the class of networks they can implement. Then, we formally prove the necessary and sufficient condition characterizing these networks, as well as the implementability of the DFT of any size power of two. All proofs are checked by the Lean kernel, and the code is openly available at: https://github.com/matteonerini/formalizing-analog-computing.

Long-horizon mobile manipulation requires a task plan and the motion that executes it. Generative planners trained on demonstration produce both in one pass, requiring neither a symbolic domain nor search. To date, however, they have relied on thousands of scripted demonstrations of fixed-base arms and executed open loop. This paper presents a hybrid flow matching planner: a single network generates the symbolic plan with masked discrete flow matching and the motion trajectory with continuous flow matching. Unlike prior generative planners, we aim to learn from a much smaller set of demonstrations and to execute the plan in closed loop. Two properties of the data compensate for the small dataset. A demonstration resumed from any of its intermediate actions is itself a demonstration, which multiplies the training samples and enables replanning after every action. Objects of the same kind are interchangeable, which turns demonstrations of one goal into demonstrations of every permuted goal. Our base implementation produces valid plans on 68% of held-out scenes; a training and generation scheme for the discrete plan raises this to 76%, and replanning after every action raises the task completion rate from 40% to 53% in a kinematic simulation. The planner matches the task completion rate of motion-only flow matching policies while additionally providing the symbolic plan, and it outperforms previous hybrid diffusion formulations on both task completion and plan validity. We validate the planner on a real mobile manipulator. Videos and project page: https://andreumatoses.github.io/research/hybrid-flow-planning

We study a one-parameter three-dimensional quadratic flow, denoted ETM-3D, on the benchmark domain $\rho\geq0$. Its homogeneous quadratic triad preserves Euclidean quadratic energy, while the full vector field has constant divergence -8. A shifted quadratic Lyapunov function yields an explicit positively invariant absorbing set and therefore global ultimate boundedness. The equilibria reduce generically to a scalar quartic, with a separate singular case at $\rho=\sqrt{5}$. Continuation reveals two saddle-node bifurcations, a generic transcritical bifurcation, a supercritical Hopf bifurcation, and a fold of limit cycles. Stable and unstable period-one branches meet at the fold with their nontrivial Floquet multipliers approaching +1. Just beyond this point, a 25-point logarithmic parameter study gives a laminar-length scaling exponent of -0.551 (95% confidence interval [-0.621,-0.481]), consistent with type-I intermittency. At larger $\rho$, a chaotic attractor coexists with an asymptotically stable equilibrium and periodic windows occur. Structural invariants are used to distinguish the flow from several classical quadratic chaotic families.

Latent world models enable robots to plan by predicting the consequences of actions. Planning long tasks with control-rate actions requires many prediction steps, which enlarges the search space and accumulates error. Skill-level actions shorten these sequences, but a symbolic skill vocabulary requires domain knowledge and labeled demonstrations. We construct skill-level actions from the inputs of a flow-matching policy trained on demonstrations segmented into complete skills. The policy maps a noise seed and an observation, optionally with a code or label, to a complete skill execution, so one execution is one world-model transition. On this mechanism we propose four action abstractions with increasing task knowledge: a compressed seed, two discrete codes learned from the demonstrations, and a symbolic label. We evaluate them with a common world-model training procedure and planning framework on simulated block rearrangement tasks that require up to 14 sequential skills. The symbolic label succeeds in over 90% of the tasks that require up to six skills and degrades beyond. Without any label, an object-centric learned code matches it on single-skill tasks and retains half to three quarters of its success on tasks of two to five skills. Ablations attribute much of the label's advantage to its planner knowing which actions are applicable, rather than to the label itself. Beyond six skills the search, not the world model, limits success. Project page: https://andreumatoses.github.io/research/flow-skill-wm

Modern models increasingly learn through black-box oracles such as humans, optimization solvers, and external tools that provide feedback without exposing their internal mechanisms. A common remedy is to learn an (action-)value function as a control variate. In this paper, we first observe that even an exact action-value function can be arbitrarily far from variance-optimal. We show that this gap arises because the value function minimizes the noise in each action's own gradient term, while an action can still affect the rest of the gradient estimator through shared parameters. A simple unbiased correction, at no extra oracle cost, can still reduce its variance by an arbitrarily large factor. Motivated by this, we then prove that the residual variance can be decomposed exactly by actions with no cross terms. This decomposition yields a closed-form variance-minimizing correction for neural-network parameters, which can be computed by a simple projection. Empirically, our correction consistently reduces the variance left by the value function and improves learning across all tasks. The source code for all experiments is available at https://github.com/Zihao-Kevin/black_box_opt.

In this paper, we develop a method that enables bimanual dexterous hands to manipulate articulated objects with a high success rate without suffering from an embodiment gap. We observe that the correlation between hand motions and object motions is dictated by the object rather than the hands and can be learned from human-object demonstrations. Based on this observation, we propose PatternDex, a method that learns this correlation and represents it as a token sequence, which we call an interaction pattern. From this pattern, PatternDex estimates the wrist motions and contact points that fit the target robot, and then trains a reinforcement learning policy that exploits these estimates as guidance. Since the guidance fits the target embodiment, the policy explores only the actions that the target robot can execute and thus achieves high success rates. PatternDex also requires only simple fine-tuning to train a new robot, since it can reuse the learned interaction pattern. We evaluate PatternDex with bimanual dexterous hands on human demonstrations from the ARCTIC dataset. PatternDex achieves, on average, a 92.8% success rate with Allegro hands, while the state-of-the-art baseline achieves 52.2%. Also, it achieves success rates above 70% with three other robot hands after fine-tuning alone. Furthermore, we verify that the learned policy transfers well to a real-world task of opening a microwave. Videos and additional results are available at https://patterndex.github.io/PatternDex/

This letter proposes DNPush, a Distributed Nonparallel Pushing force controller for cooperative object transport without inter-robot communication. Each robot uses local force measurements, a preassigned force weight, and a locally expressed desired transport direction, without requiring an object model, contact locations, a CoM estimate, or moment arm magnitudes online. DNPush regulates nonparallel force magnitudes from local force direction misalignments through a reciprocal-sine law. Under the stated contact conditions, Lyapunov analysis with LaSalle's invariance principle proves convergence to a unique rotational equilibrium, where perpendicular force contributions cancel and the net torque from normal forces is zero. Object velocity converges to the desired transport direction, and suitable contact regions exist for convex polygons. Simulations isolate the reciprocal-sine scaling mechanism, five-robot coordination, and active rotational damping. Hardware experiments with soft tactile differential-drive robots achieve a velocity direction error of 0.61 deg +/- 0.79 deg over five trials with a fixed transport direction and demonstrate object transport through waypoints with three robots.

Genomic language models can write synthetic DNA that carries no record of its origin. A generation-time watermark could provide a provenance signal, if a verifier can detect it using only the DNA sequence, a key, and published detector settings, without knowing where the generated region starts, which strand it lies on, or how it was divided into six-base tokens. We embed the SynthID tournament watermark in two genomic language models, Carbon and GENERator-v2. The verifier searches both strands, every start position, and four window lengths, and corrects its decision for the whole search. In a detection cohort of 1,544 held-out prompts, it found all 3,088 marked sequences per model, unedited and after one substituted, inserted, or deleted base. For ordinary sequences, the one-sided 95% upper confidence bound on the false-positive rate was at most 0.850%; one bound for the wrong-key control reached 1.015% (Carbon, after a deletion). In a development cohort, neither model showed a detectable change in likelihood or in predefined sequence measures. Under random edits placed without reference to the detector, detection stayed complete or nearly complete up to a 2% per-base edit rate. These results show statistical detection, not biological function, secret-key security, or robustness to an editor who sees the detector.

Objective: To assess trends in family physicians' (FPs) electronic health record (EHR) satisfaction and their experience of interoperability across organizations. Materials and Methods: Serial cross-sectional analysis of five waves of the American Board of Family Medicine's Continuous Certification Questionnaire (2022-26). Adjusted logistic regression compared odds of being very satisfied across 11 EHRs; Cochran-Armitage tests assessed vendor trends. Ease of using outside information was analyzed separately for same- and different-vendor sources. Results: Across 36,785 FPs, adjusted odds of being very satisfied relative to Epic ranged from 0.91 (95% CI, 0.74-1.11) for Elation Health to 0.17 (95% CI, 0.15-0.20) for Oracle Health/Cerner. Only Epic's users became significantly more satisfied (Benjamini-Hochberg adjusted P < .001), widening the vendor satisfaction gap. Within comparable periods, ease of using different-vendor information did not improve; fewer than 13% rated it very easy. Epic's interoperability advantage reversed by exchange type: non-Epic users had less than half the odds of Epic users of rating same-vendor interoperability very easy (adjusted OR, 0.41; 95% CI, 0.38-0.44) but 1.54 times the odds for different-vendor interoperability (95% CI, 1.35-1.75). Discussion: Rising national exchange volume has not yielded easier use of outside information. Epic's advantage is confined to its own network, so market concentration may shift, rather than solve, the different-vendor problem. Conclusion: The ABFM data are already informing EHR policy but also have utility for the EHR marketplace. While EHR satisfaction has known implications for burden and burnout, poor interoperability is a quality and safety threat that may be compounded by artificial intelligence.

Gaussian elimination with partial pivoting is the standard algorithm for solving a dense $n\times n$ system of linear equations $Ax=b$. The “growth factor” $\rho\,$ for this process may be as large as $2^{n-1}$, and when $\rho$ is large, the algorithm is unstable. Nevertheless, long experience has shown that Gaussian elimination is resoundingly stable in practice. Experiments indicate that among random matrices with independent normally distributed entries, growth factors $\rho \gg n^{1/2}$ are in a precise sense exponentially rare. Here we take a step toward confirming this observation by proving it not for the full growth factor, which depends on all the entries of the upper-triangular factor $U$, but for the corner entry $u_{nn}^{}$.