{"total":15,"items":[{"citing_arxiv_id":"2607.01275","ref_index":5,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"eXact-Prior Variational Autoencoder (X-VAE): Learning Data-Adaptive Gaussian Mixture Priors for Latent Distributions","primary_cat":"stat.ML","submitted_at":"2026-06-30T23:05:06+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":4.0,"formal_verification":"none","one_line_summary":"X-VAE uses empirical statistics from a pretrained autoencoder to set a data-adaptive Gaussian prior and introduces a latent scaling factor for controllable generation.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2606.22536","ref_index":6,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"Generative Robust Optimisation","primary_cat":"cs.LG","submitted_at":"2026-06-21T14:50:40+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"Generative Robust Optimisation defines uncertainty sets via neural network decoders over latent spaces and evaluates them with a five-point framework, validated on planning problems using Wasserstein autoencoders.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2606.18658","ref_index":8,"ref_count":2,"confidence":0.98,"is_internal_anchor":true,"paper_title":"Deep Image Prototype Learning with Geometric Heat-Kernel Priors","primary_cat":"cs.CV","submitted_at":"2026-06-17T03:55:16+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"A manifold-anchored EM algorithm selects each prototype as the highest-diffusion-centrality medoid on a heat-kernel-weighted latent graph, plus a Dirichlet regularizer, yielding sharper and more stable prototypes than Euclidean GMM baselines on cardiac and brain MRI.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2606.18132","ref_index":18,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"Knowledge Reutilization in Meta-Reinforcement Learning","primary_cat":"cs.AI","submitted_at":"2026-06-16T16:32:28+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"A meta-RL framework learns task-level knowledge on dynamics-simplified agents, organizes modes with Bayesian non-parametrics, and transfers via interfaces to heterogeneous agents, reporting 94.75-99.79% tracking error reduction with 23.8% of baseline data on locomotion tasks.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2606.12200","ref_index":85,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"Implicit Neural Representations of Individual Behavior","primary_cat":"cs.LG","submitted_at":"2026-06-10T15:19:41+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"Behavioral INR adapts INRs to behavior by mapping states to actions with FiLM-modulated episode latents for self-supervised policy inference in unlabeled data, with new policy OOD definitions.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2606.07058","ref_index":31,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"Constructing VAE Latent Spaces with Prescribed Topology","primary_cat":"cs.LG","submitted_at":"2026-06-05T08:59:55+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"A constructive framework for topology-matched VAEs on product covering space manifolds with factorized priors and closed-form KL divergences.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.29933","ref_index":2,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"CLUBench: A Clustering Benchmark","primary_cat":"cs.LG","submitted_at":"2026-05-28T13:45:37+00:00","verdict":"ACCEPT","verdict_confidence":"MODERATE","novelty_score":7.0,"formal_verification":"none","one_line_summary":"CLUBench benchmark shows conventional clustering algorithms perform comparably to deep methods, pretrained embeddings boost image/text results, and low-rank performance matrices can approximate full evaluations.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.27120","ref_index":5,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"Copula and spatial-regularized variational autoencoder for mapping disease comorbidity in West Africa","primary_cat":"stat.ME","submitted_at":"2026-05-26T14:55:51+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"A copula-integrated spatially regularized VAE is introduced to characterize geospatial comorbidity patterns of three childhood illnesses in West Africa from DHS data.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.15640","ref_index":42,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"Learning Disentangled Representations for Generalized Multi-view Clustering","primary_cat":"cs.CV","submitted_at":"2026-05-15T05:44:27+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":4.0,"formal_verification":"none","one_line_summary":"GMAE learns disentangled view-specific and view-common embeddings via dual-path autoencoders and cross-view adversarial training to boost performance on complete and incomplete multi-view clustering tasks.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.07938","ref_index":62,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Prototype Guided Post-pretraining for Single-Cell Representation Learning","primary_cat":"cs.LG","submitted_at":"2026-05-08T16:08:10+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"CellRefine adds a marker-gene-guided post-pretraining stage to single-cell models that refines the cell embedding manifold and improves downstream task performance by up to 15%.","context_count":1,"top_context_role":"dataset","top_context_polarity":"use_dataset","context_text":"In the on-domain setting, both training and evaluation are performed on single-cell RNA sequencing (scRNA-seq) data. In the out-of-domain setting, models trained on scRNA-seq data are evaluated on spatial transcriptomics data to probe zero-shot out-of-distribution generalization. We benchmark ten human single-cell transcriptomic datasets, including eight scRNA-seq datasets; peripheral blood [10, 61], pancreas [62], liver [63], myeloid [64], multiple sclerosis [65], heart [66], and lung [67], and two spatial transcriptomics datasets of liver [68]. Detailed dataset descriptions can be found in Appendix F. We report macro F1 scores for on-domain evaluations and recall@k scores for zero-shot, out-of-domain transfer tasks; recall@k denotes the proportion of test samples whose"},{"citing_arxiv_id":"2605.05813","ref_index":4,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"A Testable Certificate for Constant Collapse in Teacher-Guided VAEs","primary_cat":"cs.LG","submitted_at":"2026-05-07T07:48:41+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"For any fixed nonconstant teacher T, the best constant student has alignment cost exactly equal to the teacher mutual information I_T(X;T); a latent-only witness below this threshold with margin cannot be constant.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2604.05513","ref_index":10,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"From Unsupervised to Guided Clustering: A Variational Implementation","primary_cat":"stat.ME","submitted_at":"2026-04-07T07:10:13+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"GCVAE is a variational autoencoder that structures its latent space as a Gaussian mixture and optimizes a variational objective to make the representation maximally informative about a user-chosen guiding variable, enabling context-specific clusters.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2603.23547","ref_index":2,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"PDGMM-VAE: A Variational Autoencoder with Adaptive Per-Dimension Gaussian Mixture Model Priors for Nonlinear ICA","primary_cat":"stat.ML","submitted_at":"2026-03-20T08:54:35+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"PDGMM-VAE recovers latent sources in nonlinear ICA by using jointly learned per-dimension GMM priors that fit source-specific marginals and reduce permutation symmetry.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2509.02154","ref_index":15,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"Heavy-Tailed Class-Conditional Priors for Long-Tailed Generative Modeling","primary_cat":"cs.LG","submitted_at":"2025-09-02T10:03:10+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"C-t³VAE introduces class-conditional Student's t priors and a gamma-power divergence objective to improve class-balanced generation in VAEs under severe imbalance.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2507.05454","ref_index":15,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"Risk-Aware Aerocapture Guidance Through a Probabilistic Indicator Function","primary_cat":"eess.SY","submitted_at":"2025-07-07T20:13:48+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"A new aerocapture guidance method uses a probabilistic indicator function to estimate and mitigate failure risks, saving 71.43% to 100% of recoverable cases in high-uncertainty simulations across varied initial conditions and atmosphere models.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null}],"limit":50,"offset":0}