{"total":74,"items":[{"citing_arxiv_id":"2606.23874","ref_index":9,"ref_count":1,"confidence":0.88,"is_internal_anchor":false,"paper_title":"Identifying structural design principles shaping the computational abilities of recurrent neural networks","primary_cat":"q-bio.NC","submitted_at":"2026-06-22T19:17:25+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"Local 2- and 3-cycles enhance RNN computational capacity for Boolean functions, predicted by structural statistics, while adding interneurons boosts large networks.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2606.23840","ref_index":74,"ref_count":1,"confidence":0.88,"is_internal_anchor":false,"paper_title":"Embodied Explainability and Ontological Obstacles: Why We Struggle to Explain the Answers of Large Language Models (LLMs)","primary_cat":"cs.HC","submitted_at":"2026-06-22T18:22:53+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"An argument paper reframes LLM explainability as an embodied, situated practice based on Dourish and enactivist cognition, identifying ontological obstacles in internal explanations and advocating affordance-based designs.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2606.23478","ref_index":43,"ref_count":1,"confidence":0.88,"is_internal_anchor":false,"paper_title":"ffortissimo: A Freeform Forward-Modeling Pipeline for High-Contrast Images of Circumstellar Disks Based on Automatic Differentiation","primary_cat":"astro-ph.IM","submitted_at":"2026-06-22T15:25:22+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"ffortissimo is a JAX-based freeform forward-modeling pipeline that fits complex dust distributions and infers scattering properties in KLIP-reduced images of circumstellar disks such as HR 4796A.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2606.23204","ref_index":261,"ref_count":1,"confidence":0.88,"is_internal_anchor":false,"paper_title":"Unmasking LAION-5B: Age, Gender, Race, and Emotion Biases in Large-Scale Image Datasets","primary_cat":"cs.CV","submitted_at":"2026-06-22T11:49:23+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"Empirical audit of LAION-2B-en and LAION-2B-multi finds overrepresentation of young adults, White people, and males plus stereotypical emotion associations across two attribute classifiers.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2606.20329","ref_index":5,"ref_count":1,"confidence":0.88,"is_internal_anchor":false,"paper_title":"Constrained hybrid modelling to predict microbial dynamics and organic matter turnover in soil systems","primary_cat":"cs.LG","submitted_at":"2026-06-18T15:04:34+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"Hybrid neural-process model derives biokinetic parameters from genomic traits for soil organic matter turnover, with ecological constraints, and outperforms baselines on synthetic and real data.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2606.20299","ref_index":201,"ref_count":1,"confidence":0.88,"is_internal_anchor":false,"paper_title":"Statistical Properties of Training & Generalization","primary_cat":"stat.ML","submitted_at":"2026-06-18T14:35:53+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":2.0,"formal_verification":"none","one_line_summary":"Neural scaling laws in deep learning interact with physics constraints and inductive biases beyond classical statistics.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2606.19921","ref_index":16,"ref_count":1,"confidence":0.88,"is_internal_anchor":false,"paper_title":"eCNNTO: A Highly Generalizable ConvNet for Accelerating Topology Optimization","primary_cat":"cs.AI","submitted_at":"2026-06-18T08:17:48+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"eCNNTO applies an element-wise CNN with residual connections and final-stage training data to accelerate density-based topology optimization while generalizing across boundary conditions, loads, geometries, and mesh sizes.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2606.20751","ref_index":52,"ref_count":1,"confidence":0.88,"is_internal_anchor":false,"paper_title":"From Sentiment to Actionable Insights: A Data-Driven Public Sentiment Analysis of Advanced Air Mobility","primary_cat":"cs.CL","submitted_at":"2026-06-18T03:07:47+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":4.0,"formal_verification":"none","one_line_summary":"Applies standard sentiment classifiers and topic modeling to a large AAM discussion corpus, identifies six clusters of public concern, and lists strategies to address them.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2606.19623","ref_index":35,"ref_count":1,"confidence":0.88,"is_internal_anchor":false,"paper_title":"SEAGAN: domain-Specific and Edge-Aware Graph Attention Network for Dynamic Plant Processes","primary_cat":"cs.LG","submitted_at":"2026-06-17T22:04:24+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"SEAGAN applies a domain-specific graph attention network to classify limitation states in A-Ci curves, achieving F1-score 0.857 and accuracy 0.882 on synthetic data with known ground truth.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2606.19302","ref_index":249,"ref_count":1,"confidence":0.88,"is_internal_anchor":false,"paper_title":"Optimal scenario design for climate emulation","primary_cat":"physics.ao-ph","submitted_at":"2026-06-17T17:26:22+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"Optimizing training data via a differentiable SCM yields climate emulators that outperform those trained on six standard ScenarioMIP pathways while using less data and isolating distinct forcing responses.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2606.12610","ref_index":19,"ref_count":1,"confidence":0.88,"is_internal_anchor":false,"paper_title":"The Mathematics of AI Winters: The mathematical Taxonomy of Paradigm Fragility in AI Winter","primary_cat":"cs.LG","submitted_at":"2026-06-10T19:08:13+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":3.0,"formal_verification":"none","one_line_summary":"Established mathematical bottlenecks in representation, optimization, complexity, and high-dimensional learning aligned with the central disappointments of early AI research periods.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2606.10592","ref_index":61,"ref_count":1,"confidence":0.88,"is_internal_anchor":false,"paper_title":"Dirichlet-Guided Group Forecasting for Alleviating Over-smoothing in Time Series Forecasting","primary_cat":"cs.LG","submitted_at":"2026-06-09T08:56:21+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"DGF explicitly models multiple mode-conditioned predictive distributions via Dirichlet-guided sampling and reward optimization to preserve dynamical features in time series forecasts.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2606.10253","ref_index":3,"ref_count":1,"confidence":0.88,"is_internal_anchor":false,"paper_title":"Multi-channel Optical Vision Model","primary_cat":"physics.optics","submitted_at":"2026-06-08T23:40:15+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"Spatial multiplexing in optical neural networks is repurposed as a trainable representational coordinate, demonstrated in multi-layer architectures for image classification, regression, and hybrid vision-language captioning with over one million optical phase parameters.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2606.10170","ref_index":4,"ref_count":1,"confidence":0.88,"is_internal_anchor":false,"paper_title":"Learning Entropy and Spatial Adaptation Dynamics of Multilayer Perceptrons for Structural Point Extraction","primary_cat":"cs.LG","submitted_at":"2026-06-08T21:05:03+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":4.0,"formal_verification":"none","one_line_summary":"Spatial Learning Entropy Maps derived from MLP weight adaptations during spatial pixel prediction tasks highlight image points with high learning impact.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2606.08721","ref_index":10,"ref_count":1,"confidence":0.88,"is_internal_anchor":false,"paper_title":"A Geometric Measure of Linear Separability for Neural Representations","primary_cat":"cs.LG","submitted_at":"2026-06-07T16:31:54+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"Introduces the directional linear separability measure (LSM) as an asymmetric diagnostic for one-sided affine separability of neural representations.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2606.08477","ref_index":6,"ref_count":1,"confidence":0.88,"is_internal_anchor":false,"paper_title":"A Variability-Based Framework for Interpretable Naming in Formal and Relational Concept Analysis","primary_cat":"cs.AI","submitted_at":"2026-06-07T06:47:34+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"Presents a configurable variability-based framework for LLM-assisted naming of formal concepts in FCA and RCA, illustrated on a small pizzeria relational dataset.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2606.06174","ref_index":34,"ref_count":1,"confidence":0.88,"is_internal_anchor":false,"paper_title":"Learning to model pediatric asthma exacerbation from multiple risk factors: a case study in coastal Virginia","primary_cat":"cs.LG","submitted_at":"2026-06-04T13:47:49+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":4.0,"formal_verification":"none","one_line_summary":"A case study develops a sparse dictionary learning approach to model pediatric asthma exacerbations from multiple risk factors and reports consensus on relative risks across statistical and machine learning models.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2606.05481","ref_index":32,"ref_count":1,"confidence":0.88,"is_internal_anchor":false,"paper_title":"Towards Unified and Data-Efficient Prognostics and Health Management with Tabular Foundation Models","primary_cat":"cs.LG","submitted_at":"2026-06-03T22:00:29+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"Tabular foundation models applied to PHM via signal-to-table conversion achieve the best average ranks across prognostic and diagnostic tasks and remain competitive in low-data regimes.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2606.04033","ref_index":22,"ref_count":1,"confidence":0.88,"is_internal_anchor":false,"paper_title":"Inverse Critical Experiment Design via Gradient Optimization and a Multigroup Attention-Based Neural Network Architecture","primary_cat":"cs.LG","submitted_at":"2026-06-01T21:19:18+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"A U-Net surrogate with multigroup attention pooling is trained on OpenMC sensitivity data and combined with gradient optimization to generate grid-based critical experiment geometries that achieve c_k values up to 0.97757 for HALEU fuel validation.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2606.02341","ref_index":8,"ref_count":1,"confidence":0.88,"is_internal_anchor":false,"paper_title":"Parameter-efficient Dual-encoder Architecture with Differentiable Choquet Integral Fusion for Underwater Acoustic Classification","primary_cat":"cs.SD","submitted_at":"2026-06-01T14:49:38+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"A parameter-efficient dual-encoder model with differentiable Choquet integral fusion improves underwater acoustic classification accuracy over single-encoder baselines on DeepShip and ShipsEar datasets.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2606.00849","ref_index":1,"ref_count":1,"confidence":0.88,"is_internal_anchor":false,"paper_title":"Ultrafast formation of a large dynamic magnetic soliton","primary_cat":"cond-mat.mes-hall","submitted_at":"2026-05-30T18:49:18+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"Observation of ultrafast large dynamic magnetic soliton formation inside the linear spin-wave band in garnet films, extending tens of microns and collapsing into short-wavelength spin waves at large distances.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.30977","ref_index":57,"ref_count":1,"confidence":0.88,"is_internal_anchor":false,"paper_title":"Software Platform for Hybrid Pseudo-Random Sequence Generation and Predictability Analysis Based on LFSR and Mersenne Twister","primary_cat":"quant-ph","submitted_at":"2026-05-29T08:17:39+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":2.0,"formal_verification":"none","one_line_summary":"Software platform for hybrid LFSR-MT PRNG generation and ML-based predictability analysis, reporting inherent limitations in classical generators versus quantum randomness.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.21864","ref_index":17,"ref_count":1,"confidence":0.88,"is_internal_anchor":false,"paper_title":"A Simulation Methodology Testbed for Typhoon Sensitivity Analysis: Framework Development and Perturbation-Response Experiments with the Pangu Weather Model","primary_cat":"physics.ao-ph","submitted_at":"2026-05-21T01:21:10+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"A MATLAB/ONNX testbed integrates the Pangu AI model with PID closed-loop control to perform single-input single-output perturbation-response experiments on typhoon track and intensity.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.21352","ref_index":31,"ref_count":1,"confidence":0.88,"is_internal_anchor":false,"paper_title":"Classification of Single and Mixed Partial Discharges under Switching Voltage Using an AWA-CNN Framework","primary_cat":"cs.LG","submitted_at":"2026-05-20T16:16:51+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"AWA patterns from PD pulse amplitude, width, and area enable CNNs to classify single and mixed partial discharge sources under switching voltage with over 96% test accuracy.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.20771","ref_index":191,"ref_count":1,"confidence":0.88,"is_internal_anchor":false,"paper_title":"Cumulative Meta-Learning from Active Learning Queries for Robustness to Spurious Correlations","primary_cat":"cs.LG","submitted_at":"2026-05-20T06:14:12+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"CAML meta-learns a progressively refined inductive bias from active-learning queries to improve robustness to spurious correlations, reporting accuracy gains on minority groups across several benchmarks.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.20523","ref_index":21,"ref_count":1,"confidence":0.88,"is_internal_anchor":false,"paper_title":"Machine-Learning-Enhanced Non-Invasive Testing for MASLD Fibrosis: Shallow-Deep Neural Networks Versus FIB-4, Tabular Foundation Models, and Large Language Models","primary_cat":"cs.LG","submitted_at":"2026-05-19T21:51:02+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":4.0,"formal_verification":"none","one_line_summary":"A 354-parameter shallow-deep neural network using age, AST, ALT, platelets and FIB-4 achieved external ROC-AUCs of 0.77 and 0.67 for advanced MASLD fibrosis, slightly above FIB-4's 0.75 and 0.60 on Malaysian and Indian cohorts.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.20347","ref_index":113,"ref_count":1,"confidence":0.88,"is_internal_anchor":false,"paper_title":"Symmetrization of Loss Functions for Robust Training of Neural Networks in the Presence of Noisy Labels","primary_cat":"cs.LG","submitted_at":"2026-05-19T18:03:40+00:00","verdict":null,"verdict_confidence":null,"novelty_score":null,"formal_verification":null,"one_line_summary":null,"context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.20267","ref_index":33,"ref_count":1,"confidence":0.88,"is_internal_anchor":false,"paper_title":"Generation of Heterogeneous PET Images from Uniform Organ Activity Maps Using a Pretrained Domain-Adapted Diffusion Model","primary_cat":"cs.CV","submitted_at":"2026-05-18T21:05:28+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"A domain-adapted diffusion model synthesizes heterogeneous PET images from uniform organ activity maps, achieving high quantitative accuracy (CCC > 0.92) and visual realism comparable to real scans.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.18147","ref_index":146,"ref_count":1,"confidence":0.88,"is_internal_anchor":false,"paper_title":"Foundation Models for Credit Risk Prediction: A Game Changer?","primary_cat":"cs.LG","submitted_at":"2026-05-18T09:52:48+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"Tabular foundation models outperform standard methods in credit risk PD and LGD tasks, with larger gains on smaller datasets when used out-of-the-box.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.17704","ref_index":45,"ref_count":1,"confidence":0.88,"is_internal_anchor":false,"paper_title":"Toy Combinatorial Interpretability Models Reveal Lottery Tickets in Early Feature Space","primary_cat":"cs.LG","submitted_at":"2026-05-18T00:00:53+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"In a combinatorial toy setting, winning lottery tickets preserve families of compatible feature locations in early feature space that balance proximity to final codes with low interference, rather than specific weight subnetworks.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.17398","ref_index":32,"ref_count":1,"confidence":0.88,"is_internal_anchor":false,"paper_title":"MiniGPT: Rebuilding GPT from First Principles","primary_cat":"cs.CL","submitted_at":"2026-05-17T11:32:07+00:00","verdict":"CONDITIONAL","verdict_confidence":"LOW","novelty_score":2.0,"formal_verification":"none","one_line_summary":"MiniGPT is a self-contained PyTorch implementation of standard GPT autoregressive modeling that reaches 1.478 validation loss on Tiny Shakespeare with a 10.77M-parameter model and produces recognizable Shakespeare-style text.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.17131","ref_index":48,"ref_count":1,"confidence":0.88,"is_internal_anchor":false,"paper_title":"A Survey on Deep Learning Architectures for Point Cloud Classification and Segmentation","primary_cat":"cs.CV","submitted_at":"2026-05-16T19:37:41+00:00","verdict":null,"verdict_confidence":null,"novelty_score":null,"formal_verification":null,"one_line_summary":null,"context_count":1,"top_context_role":"method","top_context_polarity":"background","context_text":"The development of state-of-the-art (SOTA) techniques for 3D data processing has progressed in lockstep with advances in deep learning (DL) [ 47]. Since the advent of artificial neural networks, DL has matured into a foundation for landmark computer-vision architectures, including the Multi-Layer Perceptron (MLP, 1986) [79], Convolutional Neural Networks (CNNs, 1989) [48, 57], diffusion models (2015) [37, 87], and the Vision Transformer (ViT, 2021) [ 21, 101]. Although originally designed for 2D images, successive variants of these models have been adapted to ingest 3D shapes. Nevertheless, recent SOTA performance gains in 3D processing techniques appear to stem primarily from improved training protocols, such as sophisticated data-augmentation and optimization strategies, and increased model capacity,"},{"citing_arxiv_id":"2605.18878","ref_index":265,"ref_count":1,"confidence":0.88,"is_internal_anchor":false,"paper_title":"Prognostic Value of Lung Ultrasound Biomarkers for Readmission Risk in Congestive Heart Failure: A Pilot Data-Driven Analysis","primary_cat":"eess.SP","submitted_at":"2026-05-16T02:49:12+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"Pilot study uses pretrained video encoder features from lung ultrasound to predict 30-day CHF readmission, finding lower-lung views and temporal differences most informative with top MLP F1 of 0.80.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.16762","ref_index":26,"ref_count":1,"confidence":0.88,"is_internal_anchor":false,"paper_title":"Extracting redshifts from 2D slitless spectroscopic images using deep learning for the CSST galaxy survey","primary_cat":"astro-ph.IM","submitted_at":"2026-05-16T02:23:24+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"A Bayesian CNN maps 2D slitless spectral images to redshift estimates with NMAD precision 0.0104 for SNR_GI >=1 and better for brighter sources, while remaining robust to wavelength calibration errors via spatial augmentations.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.11247","ref_index":30,"ref_count":1,"confidence":0.88,"is_internal_anchor":false,"paper_title":"A Proof-of-Concept Simulation-Driven Digital Twin Framework for Decision-Aware Diabetes Modeling","primary_cat":"cs.LG","submitted_at":"2026-05-11T21:10:18+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":3.0,"formal_verification":"none","one_line_summary":"A simulation-driven digital twin framework is shown to generate interpretable diabetes trajectories for decision-aware analysis by combining benchmark data with controlled synthetic scenarios.","context_count":1,"top_context_role":"background","top_context_polarity":"background","context_text":"ically useful system must capture the evolving disease state, infer latent physiological structure, and estimate the effects of candidate interventions. Prior work has demonstrated that time-series models, including recurrent, transformer-based, and hybrid deep learning approaches, can improve short-term glucose prediction [9], [10], [23], [24], [25], [30], [31], [32]. However, predictive accuracy alone is not sufficient to meet clinical needs. The key question is not only what will happen next, but which action is most likely to lead to a safer and more desirable outcome. 2.2. Digital Twins in Healthcare Medical digital twins are commonly defined as patient- specific computational models that integrate physiological"},{"citing_arxiv_id":"2605.10178","ref_index":8,"ref_count":1,"confidence":0.88,"is_internal_anchor":false,"paper_title":"Joint sparse coding and temporal dynamics support context reconfiguration","primary_cat":"q-bio.NC","submitted_at":"2026-05-11T08:29:00+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":4.0,"formal_verification":"none","one_line_summary":"Joint sparse coding and temporal dynamics in mPFC and computational networks reduce cross-context interference and enhance separability, enabling better retention in lifelong learning without extra heuristics.","context_count":1,"top_context_role":"background","top_context_polarity":"background","context_text":"A central unresolved question is how neural ensembles remain sufficiently stable to preserve prior representations, yet sufficiently flexible to be selectively reconfigured when contexts change, thereby limiting interference during sequential learning. A related challenge arises in artificial intelligence. Despite major advances in perception and decision-making [8], standard artificial neural networks often fail when tasks or contexts are encountered sequentially, a limitation commonly referred to as catastrophic forgetting [9, 10], frequently accompanied by substantial memory and energy demands [11]. At the same time, neuroscience and artificial intelligence have long advanced through reciprocal exchange: bio-"},{"citing_arxiv_id":"2605.10142","ref_index":20,"ref_count":1,"confidence":0.88,"is_internal_anchor":false,"paper_title":"Scaling Vision Models Does Not Consistently Improve Localisation-Based Explanation Quality","primary_cat":"cs.CV","submitted_at":"2026-05-11T07:51:33+00:00","verdict":"ACCEPT","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"Scaling vision models by depth and parameter count does not consistently improve localisation-based explanation quality across architectures, datasets, and post-hoc methods; smaller models often perform comparably or better.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.05627","ref_index":41,"ref_count":1,"confidence":0.88,"is_internal_anchor":false,"paper_title":"Leveraging Image Generators to Address Training Data Scarcity: The Gen4Regen Dataset for Forest Regeneration Mapping","primary_cat":"cs.CV","submitted_at":"2026-05-07T03:28:56+00:00","verdict":"CONDITIONAL","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"Mixing real UAV imagery with 2101 AI-generated image-mask pairs improves semantic segmentation F1 scores for fine-grained forest species by over 15 percentage points overall and up to 30 points for rare classes.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.05052","ref_index":43,"ref_count":1,"confidence":0.88,"is_internal_anchor":false,"paper_title":"Interpretable Neural Networks to Predict Momentum Fluxes of Orographic Gravity Waves","primary_cat":"physics.ao-ph","submitted_at":"2026-05-06T15:49:01+00:00","verdict":"CONDITIONAL","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"Neural networks predict orographic gravity wave momentum fluxes from coarse state variables with offline R² of 0.56-0.72, learn physically meaningful relationships via SHAP, and are compared to the Lott-Miller parameterization.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.04997","ref_index":15,"ref_count":1,"confidence":0.88,"is_internal_anchor":false,"paper_title":"DualTCN: A Physics-Constrained Temporal Convolutional Network for 2 Time-Domain Marine CSEM Inversion","primary_cat":"cs.LG","submitted_at":"2026-05-06T14:58:17+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"DualTCN is the first deep-learning model for time-domain marine CSEM inversion that regresses four earth parameters, achieves high accuracy on simulated data, and runs up to 21,000 times faster than classical optimizers.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.02300","ref_index":223,"ref_count":1,"confidence":0.88,"is_internal_anchor":false,"paper_title":"A Meta Reinforcement Learning Approach to Goals-Based Wealth Management","primary_cat":"cs.LG","submitted_at":"2026-05-04T07:48:02+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"MetaRL pre-trained on GBWM problems delivers near-optimal dynamic strategies in 0.01s achieving 97.8% of DP optimal utility and handles larger problems where DP fails.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.01306","ref_index":65,"ref_count":1,"confidence":0.88,"is_internal_anchor":false,"paper_title":"Machine Learning Enhanced Laser Spectroscopy for Multi-Species Gas Detection in Complex and Harsh Environments","primary_cat":"physics.optics","submitted_at":"2026-05-02T07:28:27+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"Machine learning methods including denoising autoencoders, unsupervised interference mitigation, blind source separation, and certifiable classification are developed and experimentally validated to improve multi-species laser spectroscopy under complex conditions.","context_count":1,"top_context_role":"background","top_context_polarity":"background","context_text":"that quantifies the difference between predicted and actual outputs. Models with many lay- ers are referred to as \"deep,\" and these architectures have shown remarkable success in a va- riety of applications, ranging from noise reduction and quantitative analysis to blind source separation and molecular classification, due to their ability to model high-dimensional data and nonlinear absorption phenomena [65, 58]. 2.2.1 Regression Models for Concentration Estimation A primary application is the prediction of gas concentrations from measured absorbance spectra. Classical approaches include linear regression and partial least squares (PLS) re- 33 gression, which have been used extensively in chemometrics. For example, Su et al. applied PLS to near-infrared spectra to quantify mixture components in complex samples [66]."},{"citing_arxiv_id":"2605.00038","ref_index":52,"ref_count":1,"confidence":0.88,"is_internal_anchor":false,"paper_title":"Lottery BP: Unlocking Quantum Error Decoding at Scale","primary_cat":"cs.AR","submitted_at":"2026-04-28T15:41:05+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"Lottery BP adds randomness to belief propagation decoding and uses syndrome voting to achieve far higher accuracy on topological quantum codes while reducing reliance on expensive global decoders.","context_count":1,"top_context_role":"background","top_context_polarity":"background","context_text":"Multiple physical data and syndrome qubits are encoded into log- ical qubits with QEC codes to protect the quantum information. For example, surface code encodes physical qubits into 2D planar structure, which can be implemented easily on a superconducting quantum computer with 2D square-lattice topology, such as the very recent IBM Quantum Nighthawk QPU [52]. However, QEC codes can only protect the logical qubits, but not syndrome measurement, which extracts the parity checks as the input to the decoder. To ac- count for the measurement errors, induced by imperfect syndrome extraction, one QEC cycle includes 𝑑 measurement rounds [17], with an example shown in Figure 2. On the left is the conventional"},{"citing_arxiv_id":"2604.25617","ref_index":37,"ref_count":1,"confidence":0.88,"is_internal_anchor":false,"paper_title":"AI-Powered Surrogate Modelling for Multiscale Combustion: A Critical Review and Opportunities","primary_cat":"physics.chem-ph","submitted_at":"2026-04-28T13:24:25+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":2.0,"formal_verification":"none","one_line_summary":"A critical review of AI surrogate models for multiscale combustion that compares supervised, unsupervised, and physics-guided methods, identifies transferability and consistency challenges, and outlines future opportunities.","context_count":1,"top_context_role":"background","top_context_polarity":"background","context_text":"turbulence closures, or from boundary conditions to full flow solutions. Such mappings may be represented using a parameterized function 𝑓𝜃(𝑥), where the model parameters θ are learned from data. Deep learning architectures provide flexible approximators capable of capturing nonlinear relationships in high -dimensional datasets, making them well suited to the complexity of combustion systems [37]. Despite their flexibility, purely data -driven models often struggle to maintain physical consistency when applied outside their training domain. Consequently, the integration of physical knowledge through constraints, hybrid modelling, or governing equations has emerged as a central theme in scientific machine learning for combustion [38]."},{"citing_arxiv_id":"2604.24125","ref_index":15,"ref_count":1,"confidence":0.88,"is_internal_anchor":false,"paper_title":"Open-Vocabulary Semantic Segmentation Network Integrating Object-Level Label and Scene-Level Semantic Features for Multimodal Remote Sensing Images","primary_cat":"cs.CV","submitted_at":"2026-04-27T07:23:36+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"TSMNet uses a dual-branch text encoder and text-guided fusion module to integrate scene-level semantic and object-level label features from text with visual embeddings, achieving superior open-vocabulary segmentation on new multimodal remote sensing datasets.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2604.24004","ref_index":23,"ref_count":1,"confidence":0.88,"is_internal_anchor":false,"paper_title":"Single-Cycle Multidirectional EOG Classification Faster than Human Reaction Time for Wearable Human-Computer Interactions","primary_cat":"eess.SP","submitted_at":"2026-04-27T03:37:03+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":4.0,"formal_verification":"none","one_line_summary":"Cascaded neural networks classify 10 eye-movement classes from single-cycle EOG signals at 99% accuracy with sub-83 ms latency below human reaction time.","context_count":1,"top_context_role":"background","top_context_polarity":"background","context_text":"nal classification with wavelet and supervised learning algorithms knn, svm and dt, Sensors 23 (2023) 4553. doi:10.3390/s23094553. [21] S. Hochreiter, J. Schmidhuber, Long short-term memory, Neural Computation 9 (8) (1997) 1735-1780. [22] Y. LeCun, Y. Bengio, G. Hinton, Deep learning, Nature 521 (7553) (2015) 436-444. doi:10.1038/nature14539. [23] I. Goodfellow, Y. Bengio, A. Courville, Deep Learning, MIT Press, Cambridge, MA, 2016. [24] R. M. Noor, N. Yahya, A. I. M. Yassin, Develop- ment of an eye-controlled mobile robot system us- ing eog signals, International Journal of Electrical and Computer Engineering 8 (6) (2018) 4760 -4768. doi:10.11591/ijece.v8i6.pp4760-4768. [25] C.-T. Lin, J."},{"citing_arxiv_id":"2604.22877","ref_index":41,"ref_count":1,"confidence":0.88,"is_internal_anchor":false,"paper_title":"A Specialized Importance-Aware Quantum Convolutional Neural Network with Ring-Topology (IA-QCNN) for MGMT Promoter Methylation Prediction in Glioblastoma","primary_cat":"quant-ph","submitted_at":"2026-04-24T02:20:21+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":3.0,"formal_verification":"none","one_line_summary":"IA-QCNN applies quantum principles via ring-topology convolution and importance weighting to achieve claimed high-accuracy MGMT methylation prediction from MRI with fewer parameters and noise robustness than classical models.","context_count":1,"top_context_role":"background","top_context_polarity":"background","context_text":"data like MRI and the capacity to efficiently process large -scale datasets. CNNs are DL models developed based on the structure of the visual cortex in the human brain, utilizing a hierarchical learning approach in which each feature level is constructed upon lower-level features extracted by the previous layer, allowing the model to learn complex structures and relationships from simpl e concepts [39-41]. A typical CNN model consists of input, convolution, activation, pooling, and fully connected layers, wherein feature learning is generally performed by small filters at each layer designed to detect specific patterns. Moreover, CNN arch itectures exhibit a certain degree of robustness and invariance to spatial translations, rotations, and noise, owing to their structural characteristics [42]."},{"citing_arxiv_id":"2604.14336","ref_index":15,"ref_count":1,"confidence":0.88,"is_internal_anchor":false,"paper_title":"Mistake gating leads to energy and memory efficient continual learning","primary_cat":"cs.AI","submitted_at":"2026-04-15T18:44:23+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"Mistake-gated plasticity reduces neural network updates by 50-80% by gating changes on classification errors, improving efficiency for continual learning without added hyperparameters.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2604.13483","ref_index":4,"ref_count":1,"confidence":0.88,"is_internal_anchor":false,"paper_title":"Broximal Alignment for Global Non-Convex Optimization","primary_cat":"math.OC","submitted_at":"2026-04-15T05:14:45+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"Broximal Alignment is a novel condition under which the Ball Proximal Point Method converges to global minima in non-convex settings, generalizing quasiconvexity, star convexity, and related frameworks.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2604.12451","ref_index":10,"ref_count":1,"confidence":0.88,"is_internal_anchor":false,"paper_title":"Enhancing Laser Surface Texturing through Advanced Machine Learning Techniques","primary_cat":"cond-mat.mtrl-sci","submitted_at":"2026-04-14T08:39:27+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":2.0,"formal_verification":"none","one_line_summary":"Neural networks and random forests predict surface roughness from laser parameters and material data with high accuracy, speeding up optimization and reducing experimental effort.","context_count":1,"top_context_role":"background","top_context_polarity":"background","context_text":"parameters include laser frequency, power, and structuring speed, while the output often represents surface characteristics like average roughness. The network's accuracy is significantly influenced by its architecture, with recent trends favoring deep learning networks with hundreds of hidden layers, though this requires substantial training datasets to avoid overfitting [10]. Ensemble learning methods take a different approach by combining multiple machine learning models to produce more accurate and robust predictions than individual models. The fundamental principle is that a group of diverse models working together can outperform any single model by aggregating their predictions. Two primary approaches dominate ensemble learning: bagging and boosting."}],"limit":50,"offset":0}