arXiv:2605.14271v2 Announce Type: replace Abstract: LLM agents increasingly run inside execution harnesses that dispatch tools, allocate resources, and route messages between specialized components. However, a harness can return a correct, benign answer over a trajectory that accesses unauthorized resources or leaks context to the wrong agent. Output-level evaluation cannot see these failures, yet most safety benchmarks score only final outputs or terminal states, even though many violations occur mid-trajectory rather than at termination. The central question is whether the harness respects user intent, permission boundaries, and information-flow constraints throughout execution. To address this gap, we propose HarnessAudit, a framework that audits full execution trajectories across boundary compliance, execution fidelity, and system stability, with a focus on multi-agent harnesses where these risks are most pronounced. We further introduce HarnessAudit-Bench, a benchmark of 210 tasks across eight real-world domains, instantiated in both single-agent and multi-agent configurations with embedded safety constraints. Evaluating ten harness configurations across frontier models and three multi-agent frameworks, we find that: (i) task completion is misaligned with safe execution, and violations accumulate with trajectory length; (ii) safety risks vary across domains, task types, and agent roles; (iii) most violations concentrate in resource access and inter-agent information transfer; and (iv) multi-agent collaboration expands the safety risk surface, while harness design sets the upper bound of safe deployment.
Science Journals
arXiv:2602.12280v2 Announce Type: replace Abstract: Visual illusions traditionally rely on spatial manipulations such as multi-view consistency. In this work, we introduce Progressive Semantic Illusions, a novel vector sketching task where a single sketch undergoes a dramatic semantic transformation through the sequential addition of strokes. We present Stroke of Surprise, a generative framework that optimizes vector strokes to satisfy distinct semantic interpretations at different drawing stages. The core challenge lies in the "dual-constraint": initial prefix strokes must form a coherent object (e.g., a duck) while simultaneously serving as the structural foundation for a second concept (e.g., a sheep) upon adding delta strokes. To address this, we propose a sequence-aware joint optimization framework driven by a dual-branch Score Distillation Sampling (SDS) mechanism. Unlike sequential approaches that freeze the initial state, our method dynamically adjusts prefix strokes to discover a "common structural subspace" valid for both targets. Furthermore, we introduce a novel Overlay Loss that enforces spatial complementarity, ensuring structural integration rather than occlusion. Extensive experiments demonstrate that our method significantly outperforms state-of-the-art baselines in recognizability and illusion strength, successfully expanding visual anagrams from the spatial to the temporal dimension. Project page: https://stroke-of-surprise.github.io/
arXiv:2605.17682v1 Announce Type: new Abstract: Future 3D semantic occupancy forecasting and motion planning are central to autonomous driving, as they require models to reason about how surrounding scenes evolve and how the ego vehicle should act. Existing occupancy world models commonly discretize scenes into latent embeddings, volumetric features, or quantized tokens, and forecast future states through fixed-step autoregressive generation. This limits temporal flexibility, obscures scene evolution, accumulates errors over long horizons, and poorly matches the continuous-time dynamics of real driving scenes. We propose GEM, a Gaussian Evolution Model for non-autoregressive occupancy world modeling, where driving scenes are represented as explicit continuous 4D Gaussian primitives with learned dynamics. Instead of rolling out future occupancy states step by step, GEM directly queries the Gaussian world representation at arbitrary timestamps and splats the corresponding conditional 3D Gaussians into semantic occupancy volumes. This enables efficient forecasting over the full horizon while retaining a compact and interpretable scene representation. By decoupling spatial geometry, temporal support, and primitive motion, GEM makes the predicted world easier to inspect, as each primitive's evolution can be followed continuously over time. The same representation also supports motion planning by predicting future ego trajectories from the learned Gaussian world. Extensive experiments show that GEM achieves state-of-the-art future semantic occupancy forecasting and strong motion planning performance, while providing flexible temporal querying.
arXiv:2605.14854v2 Announce Type: replace Abstract: Human Mesh Recovery (HMR) is fundamentally ambiguous: under occlusion or weak depth cues, multiple 3D bodies can explain the same image evidence. This ambiguity is not uniform across the body, as torso pose and root structure are often relatively well constrained, whereas distal articulations such as the arms and legs are more uncertain. Building on this observation, we propose FactorizedHMR, a two-stage framework that treats these two regimes differently. A deterministic regression module first recovers a stable torso-root anchor, and a probabilistic flow-matching module then completes the remaining non-torso articulation. To make this completion reliable, we combine a composite target representation with geometry-aware supervision and feature-aware classifier-free guidance, preserving the torso-root anchor while improving single-reference recovery of ambiguity-prone articulation. We also introduce a synthetic data pipeline that provides the paired image-camera-motion supervision under diverse viewpoints. Across camera-space and world-space benchmarks, FactorizedHMR remains competitive with strong baselines, with the clearest gains in occlusion-heavy recovery and drift-sensitive world-space metrics.
arXiv:2605.14963v2 Announce Type: replace Abstract: Stereo matching on top-bottom equirectangular images provides an effective framework for full-surround perception, as vertically aligned epipolar lines enable the use of advanced perspective stereo architectures that are largely driven by large-scale datasets and monocular priors. However, the performance of such adaptations is severely limited by the scarcity of omnidirectional stereo datasets and the degradation of perspective monocular priors under spherical distortions. To address these challenges, we propose H-OmniStereo, a zero-shot omnidirectional stereo matching framework. First, we construct high-quality synthetic dataset comprising over 2.8 million top-bottom equirectangular stereo pairs to scale up training. Second, we introduce an equirectangular monocular normal estimator, specifically operating in a heading-aligned coordinate system. Beyond providing distortion-robust and cross-view-consistent geometric priors for establishing reliable correspondences in stereo matching, this design boosts training efficiency and accommodates train-test FoV mismatches. Extensive experiments show that our approach achieves higher accuracy than existing methods on out-of-domain datasets and successfully generalizes to real-world consumer camera setups using a single model. The model and dataset will be released at https://github.com/JIANG-CX/H-OmniStereo.
arXiv:2406.13187v2 Announce Type: replace Abstract: While long-tailed semi-supervised learning (LTSSL) has attracted growing attention in many real-world classification tasks, existing LTSSL algorithms typically assume that labeled and unlabeled data share nearly identical class distributions. When this assumption is violated, these methods can perform poorly because they rely on biased model-generated pseudo-labels. To address this issue, we propose a simple yet effective approach called DeCon for LTSSL with unknown unlabeled class distributions. Specifically, DeCon decouples learning into two specialized branches: a standard branch that focuses on head classes and a balanced branch that focuses on tail classes. During training, the two branches interact and gradually converge, allowing them to complement each other and ultimately achieve strong performance across all classes. Despite its simplicity, we show that DeCon achieves state-of-the-art performance on a variety of standard LTSSL benchmarks, e.g., an averaged 2.7\% absolute increase in test accuracy against existing algorithms when the class distributions of labeled and unlabeled data are mismatched. Even when the class distributions are identical, DeCon consistently outperforms many sophisticated LTSSL algorithms. Furthermore, we conduct extensive ablation analyses to tease apart the factors that are the most important to the success of DeCon. The source code is available at \url{https://github.com/Gank0078/DeCon}.
arXiv:2605.15177v2 Announce Type: replace Abstract: Test-time compute scaling is a primary axis for improving LLM reasoning. Existing methods primarily scale depth by extending a single reasoning trace. Scaling breadth by sampling multiple candidates in parallel is straightforward, but introduces a selection bottleneck: choosing the best candidate without a ground-truth verifier, since pointwise LLM judging is noisy and biased. To address this, we introduce OpenDeepThink, a population-based test-time compute framework that selects via pairwise Bradley-Terry comparison. Each generation, the LLM judges random pairs of candidates and aggregates votes via Bradley-Terry into a global ranking; top-ranked candidates are preserved and the top three quarters are mutated using the natural-language critiques produced during comparison; the bottom quarter is discarded. OpenDeepThink raises Gemini 3.1 Pro's effective Codeforces Elo by +405 points in eight sequential LLM-call rounds (~27 minutes wall-clock). The pipeline transfers across weaker and stronger models without retuning, and on the multi-domain HLE benchmark, gains appear concentrated in objectively verifiable domains and reverse in subjective ones. We release CF-73, a curated set of 73 expert-rated Codeforces problems with International Grandmaster annotation and 99% local-evaluation agreement against the official verdict.
arXiv:2602.18136v2 Announce Type: replace-cross Abstract: Fluids composed of chiral active components can exhibit odd viscosity, a property that breaks time-reversal and parity symmetries. We investigate the hydrodynamic response to monopole and dipole singularities in a compressible thin fluid layer with odd viscosity, supported by a conventional lubrication layer. Using the two-dimensional Green's function in Fourier space, we derive exact analytical solutions for the flow and pressure fields. These solutions provide a detailed description of the hydrodynamic interactions governing the motion of colloidal particles and microswimmers in confined chiral fluids, offering insight into the role of odd viscosity in modifying particle dynamics and collective behavior. The derived results are directly applicable to modeling transport, control, and self-organization phenomena in active and chiral microfluidic systems.
arXiv:2605.17684v1 Announce Type: new Abstract: While increasing research focuses on the emotional well-being of agile team members, a significant gap remains in emotion monitoring studies for Scrum Masters and meeting organizers, whose impact on team dynamics is crucial. This paper proposes a novel application integrating four carefully selected and recommended AI models to monitor the unconsciously expressed emotions of these key roles. This is achieved through: real- time transcription using a speech-to-text model; thresholding for intonation analysis to detect emotional cues in prosody; applying emotion-based vocabulary matching to identify sentiment in spoken content; and providing context-aware suggestions containing emotion keywords using an open-source, multi-module AI API. The system achieved an ASR word error rate WER of 10% in simulated meeting environments. Our evaluation shows that real- time feedback significantly improves emotion awareness during simulated agile meetings, providing Scrum Masters and meeting organizers with real-time and practical suggestions to help them quickly identify and minimize the expression of negative emotions, fostering more positive and effective team interactions.
arXiv:2605.17456v1 Announce Type: new Abstract: Multiple instance learning (MIL) is the standard approach for whole-slide image (WSI) classification and survival prediction, where attention-based models ag gregate patch features into slide-level predictions. These models treat attention weights as evidence for their predictions, but attention is optimized for classi fication, not for identifying which patches actually support the diagnosis. This conflation leads to three failures: selected patches are insufficient (keeping them alone drops Macro-F1 by 0.078), unnecessary (removing them barely changes the prediction), and unrecoverable (continuous attention scores disagree with discrete patch subsets used at inference). The central premise is that evidence quality should be optimized directly through explicit criteria- Sufficiency, Necessity, and Recov erability (S/N/R)- rather than inherited as a byproduct of classification. GCE-MIL is a backbone-agnostic wrapper implemented through three injection modes and three evidence components: a grounding mechanism that aligns selection with domain-specific concepts, noisy-OR coverage that acts as a differentiable proxy for interventional evidence search, and threshold-plus-repair recovery that converts continuous selectors into discrete subsets through marginal-guided repair. Across 9 backbones and 9 datasets (81 configurations), GCE-MIL improves average Macro-F1 by 0.024 and C-index by 0.014, reduces the continuous-discrete gap by 4-7, and increases complement degradation by 2-4. With optional tile prefiltering after discrete recovery, inference runs up to 5 faster while retaining 0.989 full-bag utility.
arXiv:2602.12755v3 Announce Type: replace Abstract: Diffusion-based image generators are promising priors for ill-posed inverse problems like sparse-view X-ray Computed Tomography (CT). As most studies consider synthetic data, it is not clear whether training data mismatch (``domain shift'') or forward model mismatch complicate their successful application to experimental data. We measured CT data from a physical phantom resembling the synthetic Shepp-Logan phantom and trained diffusion priors on synthetic image data sets with different degrees of domain shift towards it. Then, we employed the priors in a Decomposed Diffusion Sampling scheme on sparse-view CT data sets with increasing difficulty leading to the experimental data. Our results reveal that domain shift plays a nuanced role: while severe mismatch causes model collapse and hallucinations, diverse priors match or exceed well-matched but narrow priors. Forward model mismatch pulls the image samples away from the prior manifold, which causes artifacts but can be mitigated with annealed likelihood weight schedules that also increase computational efficiency. Overall, we demonstrate that performance gains do not immediately translate from synthetic to experimental data, and future development must validate against real-world benchmarks.
arXiv:2409.12190v4 Announce Type: replace Abstract: Bundle adjustment (BA) is a critical technique in various robotic applications such as simultaneous localization and mapping (SLAM), augmented reality (AR), and photogrammetry. BA optimizes parameters such as camera poses and 3D landmarks to align them with observations. With the growing importance of deep learning in perception systems, there is an increasing need to integrate BA with deep learning frameworks for enhanced reliability and performance. However, widely-used C++-based BA libraries, such as GTSAM, g$^2$o, and Ceres Solver, lack native integration with modern deep learning libraries like PyTorch. This limitation affects their flexibility, ease of debugging, and overall implementation efficiency. To address this gap, we introduce an eager-mode BA library seamlessly integrated with PyTorch with high efficiency. Our approach includes a sparsity-aware auto-differentiation design and GPU-accelerated sparse operations designed for 2nd-order optimization. Our eager-mode BA on GPU demonstrates substantial runtime efficiency, achieving an average speedup of 18.5$\times$, 22$\times$, and 23$\times$ across all benchmarks compared to GTSAM, g$^2$o, and Ceres, respectively.
arXiv:2604.12288v2 Announce Type: replace-cross Abstract: Fine-tuning is a widely used strategy for adapting pre-trained models to new tasks, yet its methodology and theoretical properties in high-dimensional nonparametric settings with variable selection have not yet been developed. We propose a source-model-augmented residual tuning (SMART) framework, which incorporates the pre-trained source model as an augmented feature into the target learner and estimates only the residual target-specific component. The approach is widely applicable, from parametric and sparse models to neural networks and blackbox machine learning models. We focus on the development of fine-tuning factor-augmented neural Lasso, resulting in SMART-FAN-Lasso. This transfer-learning framework for high-dimensional nonparametric regression with variable selection simultaneously handles covariate and posterior shifts. We use a low-rank factor structure to manage high-dimensional dependent covariates and a residual tuning decomposition in which the target function is expressed as a function of source model and other target-specific variables, thereby reducing the effective complexity of the target task. We derive minimax-optimal excess risk bounds, characterizing the precise conditions, in terms of relative sample sizes and function complexities, under which fine-tuning yields statistical acceleration over single-task learning. Extensive numerical experiments across diverse covariate- and posterior-shift scenarios demonstrate that SMART-FAN-Lasso consistently outperforms standard baselines and achieves near-oracle performance even under severe target sample size constraints, empirically validating the derived rates.
arXiv:2605.18295v1 Announce Type: new Abstract: Robust pedestrian safety is crucial to the next-generation of intelligent transportation systems. Such systems rely on active pedestrian localization and predictive collision alerts. Pedestrian localization can be supported by Ultra-Wideband technology and Bluetooth 6.0, which offer high-precision ranging and low-latency communication, making them promising candidates for vehicular collision warning systems. This paper assesses the localization accuracy of these technologies for pedestrian alerting and benchmarks their performance against Global Navigation Satellite Systems. Experimental evaluations performed in this paper focused on key performance metrics, including localization accuracy and robustness to environmental conditions. Preliminary results suggest that Ultra-Wideband and Bluetooth 6.0 can serve as viable alternatives or complements to Global Navigation Satellite Systems in certain scenarios, improving situational awareness and enabling timely pedestrian alerts.
arXiv:2409.15980v2 Announce Type: replace Abstract: Traditional machine learning-based visual inspection systems require extensive data collection and repetitive model training to improve accuracy. These systems typically require expensive camera, computing equipment and significant machine learning expertise, which can substantially burden small and medium-sized enterprises. This study explores leveraging unsupervised learning methods with pre-trained models and low-cost hardware to create a cost-effective visual anomaly detection system. The research aims to develop a low-cost visual anomaly detection solution that uses minimal data for model training while maintaining generalizability and scalability. The system utilises unsupervised learning models from Anomalib and is deployed on affordable Raspberry Pi hardware through openVINO. The results show that this cost-effective system can complete anomaly defection training and inference on a Raspberry Pi in just 90 seconds using only 10 normal product images, achieving an F1 macro score exceeding 0.95. While the system is slightly sensitive to environmental changes like lighting, product positioning, or background, it remains a swift and economical method for factory automation inspection for small and medium-sized manufacturers. The code is available at https://github.com/Yunbo-max/Cost-Effective-Visual-Anomaly-Detection-using-Unsupervised-Learning.
arXiv:2410.04626v2 Announce Type: replace Abstract: The motion of several plates in an inviscid and incompressible fluid is studied numerically using a vortex sheet model. Two to four plates are initially placed in-line, separated by a specified distance, and actuated in the vertical direction with a prescribed oscillatory heaving motion. The vertical motion induces the plates' horizontal acceleration due to their self-induced thrust and fluid drag forces. In certain parameter regimes, the plates adopt equilibrium "schooling modes," wherein they translate at a steady horizontal velocity while maintaining a constant separation distance between them. The separation distances are found to be quantized on the flapping wavelength. As either the number of plates increases or the oscillation amplitude decreases, the schooling modes destabilize via oscillations that propagate downstream from the leader and cause collisions between the plates, an instability that is similar to that observed in recent experiments on flapping wings in a water tank (Newbolt et al., 2024). A simple control mechanism is implemented, wherein each plate accelerates or decelerates according to its velocity relative to the plate directly ahead by modulating its own flapping amplitude. This mechanism is shown to successfully stabilize the schooling modes, with remarkable impact on the regularity of the vortex pattern in the wake. Several phenomena observed in the simulations are obtained by a reduced model based on linear thin-airfoil theory.
arXiv:2410.11282v2 Announce Type: replace Abstract: The Information Updating Networks (IUNs) offers significant potential for ocean exploration but encounters challenges due to dynamic underwater environments and severe system attenuation. Current methods relying on Autonomous Underwater Vehicles (AUVs) based on online reinforcement learning (RL) lead to high computational costs and low data utilization. To address these issues and the constraints of turbulent ocean environments, we propose a multi-AUV assisted data collection framework for IUNs based on multi-agent offline RL. This framework maximizes data rate and the value of information (VoI), minimizes energy consumption, and ensures collision avoidance by utilizing environmental and equipment status data. We introduce a semi-communication decentralized training with decentralized execution (SC-DTDE) paradigm and a multi-agent independent conservative Q-learning algorithm (MAICQL) to effectively tackle the problem. Extensive simulations demonstrate the high applicability, robustness, and data collection efficiency of the proposed framework.
arXiv:2411.03451v2 Announce Type: replace Abstract: The seminal work of Bencz\'ur and Karger demonstrated cut sparsifiers of near-linear size. Subsequent extensions have yielded sparsifiers for hypergraph cuts and more recently linear codes over Abelian groups. A decade ago, Kogan and Krauthgamer asked about the sparsifiability of arbitrary constraint satisfaction problems (CSPs). For this question, a trivial lower bound is the size of a non-redundant CSP instance, which admits, for each constraint, an assignment satisfying only that constraint (so that no constraint can be dropped by the sparsifier). For instance, for graph cuts, spanning trees are non-redundant instances. Our main result is that redundant clauses are sufficient for sparsification: for any CSP predicate R, every unweighted instance of CSP(R) has a sparsifier of size at most its non-redundancy (up to polylog and $1/\epsilon$ factors). For weighted instances, we similarly pin down the sparsifiability to the so-called chain length of the predicate. These results precisely determine the extent to which any CSP can be sparsified. Our result is established in the general setting of non-linear codes, or equivalently set families, yielding a VC-type theorem for multiplicative error approximation. A key technical ingredient in our work is a novel application of the entropy method from Gilmer's recent breakthrough on the union-closed sets conjecture. As an immediate consequence of our main theorem, a number of results in the non-redundancy literature immediately extend to CSP sparsification. We also contribute new techniques for understanding the non-redundancy of CSP predicates. By adapting methods from the matching vector codes literature in coding theory, we are able to construct an explicit predicate whose non-redundancy lies between $\Omega(n^{1.5})$ and $\widetilde{O}(n^{1.6})$, the first example with a provably non-integral exponent.
arXiv:2411.03936v2 Announce Type: replace Abstract: Generative modelling of multi-user datasets has become prominent in science and engineering. Generating a data point for a given user requires employing user information, and conventional generative models, including variational autoencoders (VAEs), often ignore this. This paper introduces GUIDE-VAE, a novel conditional generative model that leverages user embeddings to generate user-guided data. By leveraging shared patterns across users, GUIDE-VAE improves performance in multi-user settings, even under significant data imbalance. In addition to integrating user information, GUIDE-VAE incorporates a pattern dictionary-based covariance composition (PDCC) to improve the realism of generated samples by capturing complex feature dependencies. While user embeddings drive performance gains, PDCC addresses common issues such as noise and over-smoothing typically seen in VAEs. The proposed GUIDE-VAE was evaluated on a multi-user smart meter dataset characterised by substantial data imbalance across users. Quantitative results show that GUIDE-VAE performs effectively on both synthetic data generation and missing-record imputation tasks, while qualitative evaluations indicate that it produces more plausible and less noisy data. These results establish GUIDE-VAE as a promising tool for controlled, realistic data generation in multi-user datasets, with potential applications across domains that require user-informed modelling.
arXiv:2605.17459v1 Announce Type: new Abstract: Infrared Imaging Video Bolometer (IRVB) measures total radiation power loss from plasma in 2 dimensions through a pinhole camera geometry. Where a free-standing thin metal foil act as a broad band absorber from Soft X-Rays to IR radiation. This configuration produces line-integrated signals with poloidal and toroidal coverage that must be inverted to recover the plasma radiation emissivity distribution on a poloidal cross-section. This study compares the tomographic methods implemented to IRVB brightness data reconstruction, namely Minimum Fisher Information (MFI), Phillips-Tikhonov regularization (PTR), and Maximum-Likelihood Expectation-Maximization (MLEM). The comparison assessment is organized around several aspects of bolometer measurements, namely viewing geometry configuration, non-negativity, robustness to noise, sensitivity to prior assumptions, convergence speed, and peak preservation. The present work also details the IRVB forward modelling process, construction of synthetic phantoms, and a validation of these reconstruction methods based on typical expected emissivity profiles, namely symmetric Gaussian distribution at plasma center, symmetric hollow-radiation emissivity profile, asymmetric radiation profiles across the poloidal cross-section, and divertor-side radiation emission profiles. The outcome is to emphasize the practical tradeoffs among reconstruction accuracy, numerical stability, and suitability for real-time or offline usage of these reconstruction methods, particularly for the IRVB camera viewing system.
arXiv:2605.17461v1 Announce Type: new Abstract: Whether an interviewee's honest and deceptive responses can be detected by facial expression signals in videos has been debated and requires further research. We developed deep learning models enabled by computer vision to extract temporal patterns of job applicants' facial expressions and head movements to identify self-reported honest and deceptive impression management (IM) tactics from video frames in real asynchronous video interviews. A 12- to 15-minute video was recorded for each of N=121 job applicants as they answered five structured behavioral interview questions. Each applicant completed a survey to self-evaluate their trustworthiness on four IM measures. Additionally, a field experiment was conducted to compare the concurrent validity associated with self-reported IMs between our modeling approach and human interviewers. Human interviewers' performance in predicting these IM measures from another subset of 30 videos was obtained by having N=30 human interviewers evaluate three recordings. Our models explained 91% and 84% of the variance in honest and deceptive IMs, respectively, and showed stronger correlations with self-reported IM scores than human interviewers.
arXiv:2602.16171v2 Announce Type: replace Abstract: Directional ion transport across membranes maintains living systems in nonequilibrium, which underlies chemiosmotic energy conversion. However, the physical origin of collectively organized ion transport in primitive cellular systems remains unclear. Here, we propose a minimal model in which ion pumps collectively align through feedback between ion transport and electrostatic interactions. In the model, directional ion transport generates a membrane potential, while the resulting electrochemical potential biases pump orientation, leading to self-organized collective alignment. Numerical simulations and mean-field analysis reveal a nonequilibrium transition from a disordered state without net transport to a pump-alignment state with sustained membrane potentials. The critical behavior is consistent with the mean-field Ising universality class; however, the effective field is generated self-consistently by nonequilibrium ion transport. We further show that protocell asymmetry can bias the polarity of the membrane potential. These results provide a generic self-organizing mechanism for the emergence of bioelectricity and a physical route toward chemiosmotic coupling in protocells.
arXiv:2411.18234v2 Announce Type: replace Abstract: Cardiovascular diseases (CVDs) are any serious illness of the heart, which require accurate diagnosis to prevent fatal consequences. Hyperparameter tuning plays a critical role in optimizing machine learning model performance by selecting the most suitable parameter configurations for improved accuracy, generalization, and reliability. Grid search systematically evaluates predefined hyperparameter combinations, whereas random search samples configurations randomly from the search space enabling broader exploration with reduced computational cost. Therefore, an efficient tuning strategy is essential when developing classification models where time plays an crucial role along with the predictive capability. In this work, we propose a new hyperparameter tuning approach to tune the hyperparameters of ML models for CVD classification. The proposed random grid search combines the power of random search to explore the global space with the focused and exhaustive search of grid search in the most promising areas. This hybrid approach finds an optimal balance between exploration and exploitation and yields a robust and time-efficient ML model for classification seetings. Experimental results on state of the art models demonstrated that randomised grid search performed better than traditional hyperparameter tuning methods. In addition to the observed improvement in model performance, the computational time required for training models was substantially reduced across most of the models. Presented results of the proposed study emphasizes the reduction in training time and computational efficiency of the proposed Randomized-Grid Search method. The proposed technique has significant potential to advance ML application in healthcare providing timely and accurate CVDs diagnosis.
arXiv:2602.16990v2 Announce Type: replace Abstract: Most recommendation benchmarks evaluate how well a model imitates user behavior. In financial advisory, however, observed actions can be noisy or short-sighted under market volatility and may conflict with a user's long-term goals. Treating what users chose as the sole ground truth, therefore, conflates behavioral imitation with decision quality. We introduce Conv-FinRe, a conversational and longitudinal benchmark for stock recommendation that evaluates LLMs beyond behavior matching. Given an onboarding interview, step-wise market context, and advisory dialogues, models must generate rankings over a fixed investment horizon. Crucially, Conv-FinRe provides multi-view references that distinguish descriptive behavior from normative utility grounded in investor-specific risk preferences, enabling diagnosis of whether an LLM follows rational analysis, mimics user noise, or is driven by market momentum. We build the benchmark from real market data and human decision trajectories, instantiate controlled advisory conversations, and evaluate a suite of state-of-the-art LLMs. Results reveal a persistent tension between rational decision quality and behavioral alignment: models that perform well on utility-based ranking often fail to match user choices, whereas behaviorally aligned models can overfit short-term noise. The dataset is publicly released on Hugging Face, and the codebase is available on GitHub.
arXiv:2510.04736v2 Announce Type: replace-cross Abstract: Conditional Value-at-Risk (CVaR) is a leading tail-risk measure in finance, central to both regulatory and portfolio optimization frameworks. Classical estimation of CVaR and its gradients relies on Monte Carlo simulation, incurring $O(1/\epsilon^2)$ sample complexity to achieve $\epsilon$-accuracy. In this work, we design and analyze a quantum subgradient oracle for CVaR minimization based on amplitude estimation. Via a tripartite proposition, we show that CVaR subgradients can be estimated with $O(1/\epsilon)$ quantum queries, even when the Value-at-Risk (VaR) threshold itself must be estimated. We further quantify the propagation of estimation error from the VaR stage to CVaR gradients and derive convergence rates of stochastic projected subgradient descent using this oracle. Our analysis establishes a near-quadratic improvement in query complexity over classical Monte Carlo. Numerical experiments with simulated quantum circuits confirm the theoretical rates and illustrate robustness to threshold estimation noise. This constitutes the first rigorous complexity analysis of quantum subgradient methods for tail-risk minimization.