Papers

Automatic failure recovery is another difficult area of DevOps deployments due to the complexity of the system, high workload and the limitations of the traditional rules-based or manual approach. This study proposes a hybrid model integrating deterministic, rule-based recovery with the assistance of machine learning by fault prediction to automatically monitor failures, identify, classify and recover failures in real time. To test the architecture, a distributed log dataset of 100,000 records was used for models such as Decision Tree, Random Forest, Logistic Regression, LightGBM, Autoencoder with BiLSTM. The Decision Tree had a strong performance in defect identification with an F1 score of 80.8%, recall of 80.0%, precision of 81.6%, and accuracy of 89.4%. An impressive 83.3% success rate was achieved by the automated recovery measures, resulting in a 94.6% decrease in mean time to recovery (MTTR), a 95.4% reduction in downtime, and annual savings of 7,462. The combination of explainable rule-based logic and adaptive AI prediction in the framework means it is more resilient, operationally efficient, and scalable, and an effective, autonomous approach to DevOps environments evaluated under controlled experimental conditions.

For the unicritical family $f_{c,j}(z)=z^j+c$, the Multibrot sets $M_j$ converge geometrically to the closed unit disk as $j\to\infty$, but this limit does not quantify their finite-degree area behavior. We estimate Area($M_j$) using escape-time pixel counting while separating the effects of the iteration threshold $K$ and spatial sampling density $\rho$. The finite-iteration sets form a nested decreasing sequence, while grid refinement is treated empirically. Across $K\in\{25,100,500,1000\}$ and $\rho\in\{128,256,512,1024\}$, the final grid refinement changes the reported areas by at most $1.1\times10^{-3}$. At $K=1000$ and $\rho=1024$, the estimates are well described by $C(j)\approx\pi-2.49216j^{-0.57103}$. Independently, Laurent coefficients of the exterior conformal map and Gronwall's area formula yield finite-truncation upper bounds $U_{1000}(j)\approx\pi-2.07405j^{-0.643427}$. Across all tested degrees, the grid estimates remain below the corresponding conformal bounds, and both sequences exhibit similar power-law decay of the area deficit from $\pi$.

Agentic AI systems are increasingly adopting automated pipelines that integrate multiple tools. While prior research have studied task success and task completion of such agentic systems, research on agent to tool interaction chain via a wrapper and API, specifically in biological agentic workflows, remains limited. This study investigates specific failures in agent to tool interaction where a tool invocation appears successful but some or all of the information or functionality from the tool via its API / wrapper is incomplete or missing and there are no communications to the user or the agent regarding the missing information. We call these 'silent failures' because the user or the agents are not aware that such a failure has occurred. For the purpose of this study, we developed an audit mechanism to identify such silent failures in agent to tool interaction chains, by examining 15 scientific tools (and their associated API documentation and tool documentations) integrated within the ToolUniverse environment (ToolUniverse serves as our experimental environment rather than the object of the study itself). We structure our study around 7 failure locus, characterising where the failure occurs in the chain. We observed 91 failures (manually validated post LLM based candidate discovery and automated testing), most frequent of them being missing data or fields and inconsistencies in search, filtering or ranking criteria. Most of the 91 failures occurred in API layer (52) or wrapper layer (25), with a potential of silent failure amplification downstream. The results show that silent failures originate upstream of the event and propagate downstream into apparently valid scientific outputs. We propose a concept of contextual reliability to handle such failures and suggest mechanisms for testing, disclosing, monitoring, and measuring such failures across the agent-tool interaction pipeline.

COBOL remains widely deployed, yet representative corpora reflecting real production code are rarely available, limiting rigorous benchmarking of modernization approaches. We propose a systematic agentic AI pipeline for generating realistic COBOL programs, combining specification-driven generation with iterative degradation guided by patterns and complexity targets extracted from real production code. Here, realism is understood as structural fidelity to production code as captured by our metrics. We evaluate whether degradation reaches target complexity levels while preserving business behavior, and examine the limits of the approach, across three programs from distinct business domains. Results show the pipeline reliably produces syntactically valid programs and moves them toward realistic structural complexity. However, preserving business behavior is not always achieved by construction, and targeting structural metrics independently of business logic risks producing programs whose complexity does not reflect a plausible maintenance history. We discuss these limitations and outline a more realistic alternative as a direction for future work, generating legacy programs from scratch along a simulated development history.

We review the conceptual and theoretical status of twin-universe models, from Sakharov's early ideas on baryogenesis, CPT symmetry, and reversed arrows of time to recent quantum-cosmological scenarios involving pair-created entangled universes. Particular attention is given to the distinction between Sakharov-type cosmological symmetry, modern bimetric theories of gravity, and speculative models involving negative mass sectors. We summarize the main consistency conditions faced by bimetric gravity, including the Boulware-Deser ghost, Higuchi-type bounds, covariant conservation laws, and no-go results for positive- and negative-mass sectors. We then examine the Janus cosmological model and related black-hole replacement proposals, emphasizing the difference between heuristic twin-universe narratives and fully specified relativistic field theories. We argue that while twin-universe ideas remain a fertile framework for questions about the arrow of time, matter-antimatter asymmetry, and the quantum origin of the universe, models combining bimetric gravity, negative mass, dark-sector phenomenology, and alternatives to Kerr black holes face a substantial burden of mathematical and observational proof.

The topological classes of coherent structures in magnetohydrodynamic turbulence are studied using the irreducible representations of the SO(3) and SO(2) rotation groups. We probe the coherent structures around axisymmetric magnetic fields in the universe. Such axisymmetric fields can be found around filamentary structures as well as cosmic strings. We have considered a Gaussian and a non Gaussian magnetic field. We find that the field distribution leads to different coherent structures for the two cases. Our results show that the scattering of particles from the Gaussian structure depends predominantly on two lengthscales. In the case of the non Gaussian distribution, we can define a third intermediate lengthscale, similar to the Taylor lengthscale. In both the cases, we find that there is a dominance of the polarization anisotropy which indicates that the field gradients are forced into rigid spatial configurations. In the early universe, this will change the mean free path of the baryons and leptons in the coherent structures. Trapping of charged particle as well as their acceleration will be more frequent in the case of a magnetic field with a heavy tailed distribution. The lengthscale of density inhomogeneities close to these structures will thus be determined by the magnetic field distribution of the fields.

For a $3$-uniform hypergraph $H$, let $\sigma(H)$ be the minimum over the vertices of $H$ of the spectral radius of the link. Lin, Lu, Yuan and Zhao conjectured that $\sigma(H)>\tfrac{2n}3-2$ forces a perfect matching in a $3$-graph of order $n$ divisible by three, and Lu and Yuan conjectured that $\sigma(H)>\tau(q)$ forces one in a $q$-balanced $3$-partite $3$-graph, where $\tau(q)=\sqrt{q(q-1)/2}$ for odd $q$ and $\tau(q)$ is given by a quartic for even $q$. In this paper, we prove both perfect matching conjectures for large order. Each proof passes through a stability theorem for fractional matchings: a fractional vertex cover of deficient weight with a zero coordinate is obtained, the link of that vertex carries the induced cover, and the spectral radius of such a link is bounded by an inequality whose extremal cases identify the corresponding space barriers.

We derive a Lipatov type effective action for the Reggeized gravitons. In place of the order by order construction of the effective currents from diffeomorphism invariance, we propose a current which is a single ordered exponential $W_{ij}$ along the light cone, generated by the optical tidal matrix integrated once. The nested ordering of it reproduces the known perturbative answer with some new terms added. These new terms are non-local along the light cone and have a structure which the diffeomorphism construction cannot produce, the form of the third order contribution to the current is also calculated. The effective action, the current to Reggeon vertex and the current to current interaction are obtained in closed form, in three equivalent writings, one of them covariant. Generally defined $W_{ij}$ reduces to four parameters of the ray bundle, its density, its shear, the orientation of the shear and a rotation of the image. The effective current uses three of them, being the second light cone derivative of a single scalar, it is not sensitive to the orientation. Comparing with the standard eikonal gravitational Wilson line, which is the phase, $W_{ij}$ is the Hessian of the same eikonal: it carries the shear and the focusing that the Abelian phase does not, and it is inverse to the Jacobi propagator of the congruence at first order in the field. A generalization of the proposed ordered operator is discussed as well as applications of the gravitational effective action.

Simulation success is not equivalent to structural correctness for LLM-generated circuits. We define and measure four evaluation levels – schema validity, topological validity, backend executability, and component-set agreement – on a 150-circuit trilingual benchmark, through a deployed pipeline built on a typed circuit interchange representation. The levels are not nested. On gpt-4o-mini, 16 of 150 circuits (10.7%, 95% CI 6.7-16.6) were rejected by the topological validator but executed in ngspice with no error or warning; 12 of these contained exactly the requested components, with one terminal disconnected. Conversely, 7 circuits (4.7%) passed the validator and ngspice refused them. Ten failed both checks and 117 passed both, so each check detects a class the other misses. A minimal three-component divider shows the cost: a dangling resistor reports 5.00 V instead of 2.50 V while ngspice stays silent. A paired ablation, in which every arm is evaluated from the same model sample rather than a fresh one, separates each repair stage from sampling noise. On a stratified 45-circuit subsample, model repair raised topological validity from 40.0% to 84.4% (+20 circuits, no regressions) while moving executability by a net 6 (+7, -1), an effect this sample size does not resolve, and component agreement by 2. One circuit moved in opposite directions at two levels in a single repair step. Against a direct-netlist baseline the pipeline executed 88.7% against 47.3%, or 62.7% under an accounting that credits the baseline with every failure we cannot confidently attribute to the netlist. These results support a narrow methodological conclusion: structural validation and simulation should be reported as distinct evaluation stages for LLM-generated circuits. A circuit that runs is not necessarily structurally valid, and a structurally valid circuit is not necessarily executable.

Rapidly growing AI workloads are driving large investments in AI datacenters. This survey classifies industrial AI accelerators into four architectural categories and compares their compute and memory organizations. It examines how node-, rack-, and pod-scale interconnects support collective communication, and traces architectural evolution across accelerator generations. The analysis connects advances in arithmetic throughput with changes in precision, data delivery, execution coordination, communication, power delivery, and cooling. It also discusses future design challenges arising from workload diversity, data movement, infrastructure constraints, and model evolution, showing how the trade-off between generality and specialization extends from individual accelerators to datacenter-scale systems. GitHub: github.com/Yufeng98/AI-datacenter

With sparse and event-driven computation, spiking neural networks show great potential for achieving accurate and power-efficient UAV visual tracking. However, existing SNN-based trackers typically use spike firing rates only for power consumption and lack explicit optimization of actual spike activity. Moreover, regulating spike activity alone does not explicitly encourage stable target representations under partial observations and temporal appearance changes. We propose SBMVTrack, a fully spiking tracking framework that combines spike activity regulation with complementary multi-view representation learning. Specifically, SBMVTrack introduces Energy-Weighted Spike Budgeting (EWSB), which incorporates layer-wise computational costs when regulating spike firing rates and penalizing saturated activations, thereby reducing redundant spike computation. To further improve target representations under the spike budget constraint, we introduce Masked Multi-View Target Modeling (MVTM), which treats the initial template, online template, and search region as temporal views of the same target. By aligning target embeddings between masked and corresponding unmasked views and enforcing cross-view identity consistency, MVTM encourages robustness to missing local cues and temporal appearance changes. Experiments on four UAV benchmarks demonstrate competitive tracking performance with a 24.1% reduction in estimated power consumption relative to the baseline. On VisDrone2018, SBMVTrack achieves a success rate of 70.0%, exceeding SpikeTrack by 9.7 percentage points while reducing estimated power consumption by 45.7%. The source code will be released upon acceptance.

Assigning a standard commodity code to a free-text purchase line underpins enterprise spend analytics, and its reported accuracy cannot be checked. Published results use proprietary data or private samples under undocumented protocols; no two compare and there is no public benchmark. We build a harness with four protocols over 1.26 million labelled purchase lines from two US state governments. Enterprises rebuy continuously, so random splits put matching item text on both sides: 60.7% and 61.2% of test rows, 54.7% and 60.6% byte for byte. On California orders the best classical baseline scores 56.0% on repeated text but 31.9% on novel text, a 24-point gap widening with taxonomy depth. Embedding retrieval shrinks the gap from 23.8 to 19.1 points; a fine-tuned transformer does not escape it. On one corpus leaky evaluation cannot separate retrieval from the transformer, while three leak-free protocols put the transformer 2.3 to 3.6 points ahead, so leaky splits hide real differences, not just flatter both. We derive an upper bound on text-only classifiers from identical text with conflicting codes: 79.4% commodity accuracy on one corpus against 98.4% on the other, so accuracy does not compare across datasets. Spend-weighting the bound reflects the amount field more than labels. In a deployment of 920,927 lines, reviewers accepted 32.4% of suggestions on the novel-text population; 29.7-35.2% brackets benchmark rates near 31.9% and 34.8%. We release the harness in production use.

We present the first method for native, continuous gradient descent for machine learning models with $p$-adic parameters. Existing native optimizers are discrete, mostly combinatorial searches, as the $p$-adic numbers $\mathbb{Q}_p$ are totally disconnected, with standard losses that are flat away from their minima. To enable continuous optimization, we propose working with $\mathbb{Q}_p$ via its Berkovich affine line: a canonical, path-connected expansion of $\mathbb{Q}_p$ that preserves its isometries and uniquely extends its analytic maps. This hull is a metric tree with interpretable points and local derivatives, which we show enables effective optimizers and backpropagation. We formulate gradient descent and show that its approximations efficiently learn linear models with coefficients in $\mathbb{Q}_p$ to do modular arithmetic, an XOR-like task not expressible by linear models in $\mathbb{R}$. We also demonstrate momentum and Adam variants, linear regression, and classification on binary-encoded hierarchies (Quillian semantic networks), addressing open problems posed by Martins (2025). Library at https://github.com/google-deepmind/padic-ml

Training data quantity and quality greatly affect object detection model performance, regardless of model architecture. For object detection in deep-sea video and imagery, where the objects of interest (primarily organisms) are sparse, faint, and hard to identify, incremental improvements to detector performance may require an iterative approach to data labeling and management. This paper presents mbariml, a Python-based video and image analysis pipeline built around the data labeling and management process. mbariml uses an Ultralytics YOLO detection model, runs it over still images or video, stores every detection as a reviewable region of interest, groups those regions by visual similarity so that a human can accept or reject them in bulk, and exports the result as training data, statistics, image sidecars, and additional metadata. Existing YOLO and Pascal VOC datasets can be imported into the same database, so a legacy training set can be reviewed, extended, and re-exported alongside new detections. The human review stage is the center of the design: an annotator can validate, relabel, resize, delete, and draw entirely new localizations, optionally assisted by the SAM3 segmentation model, and every one of those edits is written back to the same database the detector wrote to. Video receives particular attention: the software treats each tracker-produced track as a provisional observation and selects one representative frame instead of retaining every detection in the track. We describe the pipeline stage by stage, including how each track's representative frame is chosen (from a user-selected third of the track, the middle by default), which we examine on 684 tracks from seafloor video.

We prove that the principal minors of an orthogonal projection matrix determine the coordinatewise absolute value of its image. More precisely, if $W_1,W_2\subset \mathbb{C}^n$ are subspaces whose orthogonal projection matrices have equal corresponding principal minors of all orders, then $|W_1|=|W_2|$, where $|W|$ is the image of $W$ under the coordinatewise absolute value map. The proof is divided into different cases related to the connectivity of the associated matroids. We provide a stratification of the Grassmannian in terms of the connectivity levels and give a structure theorem for the elements of each stratum.

Deploying Large Language Models for runtime operational triage incurs prohibitive latency (>100-500 ms), high VRAM requirements (>4-8 GB), and excessive energy dissipation. Extending Mandelbrot Fractal Neural Synthesis (Dagli et al., 2026), this paper presents the Universal Fractal Natural Language Decision Map, realized via the werr machine-native edge reflex runtime and the production answerr platform (https://answerr.me). Operating entirely without stored weight tensors (0 Bytes VRAM), the engine synthesizes deterministic decisions—noul (Boolean), choice (categorical), and score (ordinal)—by dynamically modulating 24-byte coordinate seeds along the chaotic boundary of the Mandelbrot set and evaluating multi-scale escape dynamics. Drawing inspiration from biological System-One reflex arcs, the engine introduces: (i) an Auto-Seed Router with domain projector Phi_D yielding a +28.8% accuracy gain over linear baselines; (ii) an Information-Theoretic Semantic Token Damping Filter (T_desc = 0.045) insulating against prompt injections (0.0% empirical bypass; 95% Wilson CI: [0.0%, 27.8%]) while pruning iterations by 45.8% (accelerating throughput 2.5x to 3.31 ms latency); (iii) a Multi-Scale Harmonic Tripod Fusion; (iv) a Coupled Margin Expansion Operator (Pitchfork Bifurcation Offset); and (v) a Cyclic Z/9Z Modular Resonant Grid Discretization based on the closed sub-ideal {0,3,6} (Lean 4 Mathlib ZMod 9), reducing FLOPs by 68.4%. Evaluated on JevBench (N=231), werr achieves 100.00% TypeSafe compliance and 81.65% calibrated accuracy with 7.08 ms median latency. We provide an OpenAI-compatible API and demonstrate deployment on 32-byte EVM smart contracts via the open-source werracle on-chain oracle (21,438 gas).

We compare the Feng–Huang and covering definitions of weighted topological entropy for equivariant continuous maps between dynamical systems. The two quantities agree on the whole space by their variational principles, but need not agree on non-invariant subsets. We prove that the Feng–Huang entropy is bounded above by the covering entropy on every subset, without assuming invariance. The proof does not rely on measure theory. Our main result provides a new perspective on the two relative weighted variational principles.

In this paper, we motivate the need for a participatory speech dataset creation framework through a case study of the LGBTQIA+, or queer, community - a community with documented concerns about AI and reported harms, including attempts to develop 'gaydar' technologies that purportedly identify individuals as queer. We review common speech data collection practices, why these methods may be unsuitable for engaging with queer speakers, and discuss previous efforts in participatory AI with queer community engagement, as well as participatory endeavours specific to speech data collection for other marginalized communities. From this review, we develop a conceptual framework for participatory speech data curation by, for, and with marginalized communities drawing on insights from co-design and knowledge sharing. We propose a framework comprising overlapping and two-way processes of defining a community, project formulation, modes of participation, and personal autonomy.

Assuming the nonexistence of uncountable measurable cardinals, we construct an $\mathbb F_2$-algebra and a covering class of right modules which is not closed under colimits of countable chains. This gives a counterexample to Enochs' conjecture under $V=L$.

We investigate the sampling efficiency in a segment-based continuous-time hybridization-expansion quantum Monte Carlo solver within dynamical mean-field theory for the doped SU(3) Fermi-Hubbard model. Focusing on the low-temperature and strong-coupling regime, where a ferromagnetic state appears upon hole doping away from one-third filling, we compare several update schemes. Adding the double-flip update to the basic local update yields the smallest integrated autocorrelation time of the majority-flavor occupation in the ferromagnetic regime, whereas adding the flavor-permutation update is the most effective in the paramagnetic regime. We further clarify the origin of this difference by analyzing the characteristic occupation changes caused by accepted updates.

Topological edge states offer a robust route to photonic devices and systems that remain functional despite structural imperfections. Topological waveguide arrays have been shown to deliver robust coupling, but only for full power transfer between the two edge waveguides—a robust route to an arbitrary intermediate ratio has remained unknown. Here we show that the topological edge states of a finite Su-Schrieffer-Heeger chain, realized with only four identical waveguides, form an arbitrary-ratio power splitter whose ratio is robust to both the device length and fabrication errors. Two near-degenerate edge supermodes beat to produce a step-like power transfer with a flat, length-insensitive plateau, in contrast to the sinusoidal transfer of a conventional directional coupler. The splitting ratio at this plateau is governed by the difference between the two waveguide gaps, making it immune to correlated fabrication offsets. We derive this behavior analytically and confirm it numerically and experimentally on a conventional silicon-on-insulator platform. Beyond this device, our analysis can be applied to general topological waveguide arrays, while the robust splitter offers a practical building block for fabrication-tolerant photonic circuits.

Large vision-language models (VLMs) can be efficiently deployed under stringent memory and latency constraints through post training quantization (PTQ). However, most PTQ methods are designed for unimodal large language models (LLMs). These methods treat quantization errors as isotropic perturbations under the Euclidean assumption, which provides weak guidance on directions most sensitive to quantization in VLMs. Consequently, directly adapting unimodal PTQ approaches or solely employing modality-specific scaling often leads to uneven bit-width distribution and inconsistent performance in low-bit settings. To address these challenges, we propose Riemannian Geometry-Sensitive Quantization (RGSQ), which formulates quantization as a reconstruction problem under a unified Fisher-Riemannian metric. RGSQ identifies modality-specific sensitive directions via Riemannian manifold mappings built from modality-partitioned empirical Fisher factors and fused into a modality-aware Kronecker-structured metric. We then apply geometry-aligned rotations to reorient the local tangent frame, steering low-bit perturbations toward loss-insensitive axes. Finally, we apply a whitening transformation that maps the Riemannian objective to an equivalent Euclidean form, enabling standard unimodal PTQ methods to evaluate multimodal quantization error under their original assumptions. Across an extensive and diverse set of mainstream VLM benchmarks, RGSQ achieves the highest accuracy and stability under extremely low-bit settings (W2A8 and W3A8). It outperforms VLM-aware baselines, such as MBQ and MQuant, by up to 5.9% and surpasses single-modality improvements by up to 8.6%.

In this paper, we examine speech datasets for their inclusion of LGBTQIA+, or queer, voices and provide a taxonomy of tensions to better understand why there is a lack of such voices in current speech technology datasets. Through an audit of six diverse speech datasets, we find that measurable queer representation is low (0-1.4% of speakers) - insufficient for robust disparity measurement. We take this community as a case study to consider what challenges and tensions are associated with collecting speech data from marginalized communities. For comparison, we audit an additional two datasets from the speech sciences that were created by, for, and with the queer community. We note that many customs in speech dataset collection efforts in AI and speech technology research may conflict with values emphasized in participatory approaches with marginalized communities, and provide a taxonomy describing these tensions.

Consistent multi-view object segmentation is critical for 3D perception and robotics, yet remains challenging under severe viewpoint and occlusion changes. Existing methods typically perform 3D instance segmentation on point clouds or rely on offline 2D mask-matching pipelines. However, 3D instance segmentation is limited by scarce 3D annotations, while offline 2D matching suffers from object identity ambiguity across frames. To leverage strong 2D and 3D priors jointly, we propose SAM-V (Geometry-Aware Segment Anything for Multi-View Instance Segmentation). Instead of combining the two priors through post-hoc matching, SAM-V directly integrates features from a feed-forward geometry model (VGGT) into a 2D segmentation foundation model (SAM), trained end-to-end for cross-view instance prediction. SAM-V introduces a prompt-fusion mechanism that enriches sparse SAM prompt tokens with view-specific camera tokens and local VGGT features, making the prompt representation both view-aware and spatially grounded, together with a mask decoder that attends to dense 2D and 3D features. By conditioning the mask decoding directly on multi-view geometry, SAM-V produces consistent multi-view segmentation of a prompted object in a single forward pass without offline mask matching or explicit 3D reconstruction. On the IGGT 3D tracking benchmark, where consistent instance identity across frames directly determines performance, SAM-V improves overall IoU by 5 points and frame-level recall by 12 points on the ScanNet++ split over the state-of-the-art multi-view instance segmentation baseline and leads on all metrics in the zero-shot ScanNet split. Our code and pretrained models are available at https://github.com/gong208/SAM-V.git.

In this paper, we develop a high-order variable-scaled energetic variational neural network (VS-EVNN) method for phase-field gradient flows. Each discretization stage uses an updated-state network whose spatial inputs are centered and divided by the interface scale, while the sampled energy and movement penalties remain unchanged relative to the baseline EVNN formulation. This rescaling is motivated by a coordinate change in the energy that normalizes the gradient coefficient of the transformed functional. We combine the scaled representation with high-order, multi-stage variational extrapolation schemes and a heuristic adaptive time-stepping controller. Each stage is solved as a network minimization problem, in which the earlier stages serve as fixed anchors on the quadrature grid, yielding a neural realization of these multi-stage variational schemes, including a mobility-weighted, mass-conserving version for Cahn–Hilliard flows. The method is tested on one- and two-dimensional Allen–Cahn and Cahn–Hilliard flows, the latter in the mass-conserving metric. Under the reported optimization settings, the scaled networks yield lower solution errors than the unscaled EVNN baseline, and the higher-order schemes improve the accuracy in the test examples.

Interstellar Objects (ISOs) may provide direct evidence of planetesimals ejected from early-stage solar systems. Studying these objects and predicting their detection through various surveys is vital for enhancing our understanding of planetary formation. We show how ISO size distribution (SFD), f_D~D^-q, controls expected detections in current and future surveys. We focus on TESS and the Vera Rubin Observatory LSST while analyzing Pan-STARRS and ATLAS. TESS is expected to detect ~1 ISO from shallow SFDs (q<2.5). On the other hand, LSST is likely to detect ~0.2-100 ISO yr^-1 over its lifetime. ATLAS is likely to detect one non-cometary object for populations with q<3. We further estimate ~0-8 detections of cometary like-3I/ATLAS objects by LSST in 10 years and ~1 cometary object detected by TESS over its mission time, independent of the nucleus size distribution. By considering both 1I/'Oumuamua and 3I/ATLAS from the same SFD, our results suggest that more asteroid-like objects are produced rather than comets during the early stages of planetary formation. Imminent detections of ISOs by LSST will help constrain the slopes of the SFDs for asteroid-like and cometary ISOs better, and, if applicable, differentiate between them.

Models that predict which Google Play apps will be removed are trained on labels that record only whether an app was still in the store at a later observation. A disappeared app is labeled removed and a present one stable, but neither records why. A voluntary withdrawal and a policy takedown both produce removed, and an uncaught spam app keeps stable. This work calls that mismatch label noise. Three detectors from different methodological families are applied to the 870,514 apps of Mohsen, Karastoyanova, and Azzopardi (2022): Isolation Forest, flagging apps unusual in the feature space, Neighborhood Disagreement, flagging apps whose nearest neighbors carry the opposite label, and Prediction Inconsistency, flagging apps a classifier labels differently from the data. The apps flagged by all three, the overlap, number 7,598 at default settings and are the strongest mislabeling candidates. Two questions follow. First, does removing flagged apps improve the model? It does not. No detector, overlap, or union beats the baseline, and the loss grows with the number removed. Second, do flagged apps appear less often than expected among apps whose label VirusTotal and Quark Engine confirm? Among confirmed removals they do, falling to 0.43 times the expected rate as the threshold tightens, while an excess on the stable side disappears once the age of the scanned apps is accounted for. A model trained on only the 3,021 trainable overlap apps reaches a test AUC of 0.2518, far below chance, so the relationship between features and labels there runs opposite to the rest of the data. The flagged apps run wrong in both directions: abandoned apps that resemble spam carry stable, while apps that look healthy carry removed. The value of the detectors lies in characterizing this label noise. They locate a small set of candidates they cannot profitably remove.

Forceful manipulation is challenging for humanoid robots because interaction forces can disturb whole-body balance. We introduce the Supporting Hand Strategy (SHS), which enables a humanoid to brace against the environment with one hand while performing forceful manipulation with the other. SHS optimises a task-conditioned support configuration that guides two synchronous reinforcement-learning policies, without human motion data or online whole-body trajectory planning. On a Unitree G1, SHS achieved usable contact forces up to 60 N, compared with a maximum sustained force of 13.5 N without environmental bracing, while substantially improving force tracking over a task-independent support configuration. The same policies generalised to different task regions without retraining. SHS therefore provides a simple mechanism for substantially extending humanoid forceful-manipulation capability.

We study how timestamps affect dynamic disclosure. At a random date, a sender privately obtains hard evidence about an evolving binary state and chooses when to disclose it. With timestamps, the unique equilibrium features immediate good-evidence disclosure and timestamp-dependent bad-evidence delay. Without timestamps, high priors relative to impatience generate a stock of undisclosed good evidence, which is stochastically purged before permanent transparency. High impatience yields immediate good-evidence disclosure, while bad evidence can be arbitrarily delayed and disclosed in bursts. Timestamps prevent pretending old good evidence is fresh and certify bad evidence is old, accelerating good-evidence disclosure and facilitating bad-evidence disclosure.

Smart meter (SM) data provides fine-grained visibility into household energy consumption, but also exposes users to privacy risks. Inference attacks, known as non-intrusive load monitoring (NILM), can perform appliance-level inference from aggregate signals and recover sensitive behavioral patterns. In practice, attacker models are unknown and heterogeneous, making robust defense challenging. We formulate SM privacy protection as a black-box inference defense problem, aiming to reduce the recoverability of appliance-level information while generalizing across diverse and unseen attackers. We propose a proxy-guided hierarchical reinforcement learning framework that learns battery-based load-shaping policies to inject realistic but misleading appliance-level signatures into the aggregate signal, thereby disrupting the structured patterns exploited by NILM. A self-supervised aggregate-structure privacy probe provides a reconstruction-error-based surrogate reward for disrupting recoverable load structure, while a signature library makes the perturbations appliance-relevant and physically realizable through battery control. We provide theoretical rationale showing that proxy-guided optimization improves inference robustness under attacker diversity. Experiments on real-world datasets UK-DALE and REDD demonstrate strong cross-model and cross-appliance generalization. Across six unseen NILM attackers, covering four appliances on UK-DALE and five on REDD, our proposed defense increases average appliance-level RMSE by 107% and 166%, respectively, while reducing F1 score by 79% and 80%.