{"total":74,"items":[{"citing_arxiv_id":"2606.28207","ref_index":31,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"Three-Body Earth-Moon Transfers with Different Departure/Arrival Orbital Altitudes: New Phenomenon and Diffusion Model-Augmented Construction","primary_cat":"math.OC","submitted_at":"2026-06-26T15:58:37+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"Identifies discontinuous TOF behavior in three-body Earth-Moon transfers and augments grid search with a diffusion model, reporting 47-56% better convergence and 39-40% time savings for different orbital altitudes.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2606.10023","ref_index":28,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"Learning the Universe: Posterior Reliability of Neural Generative Models in High-Dimensional Field-Level Inference of Cosmic Initial Conditions","primary_cat":"astro-ph.CO","submitted_at":"2026-06-08T18:08:00+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"Generative models for cosmological field-level inference can reproduce posterior means and cross-correlations yet fail to capture correct uncertainty geometry when validated against HMC reference samples.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2606.00295","ref_index":64,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"Adaptive Order Policies for Masked Diffusion","primary_cat":"cs.LG","submitted_at":"2026-05-29T19:26:53+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"A policy network learns to choose unmasking order in masked diffusion by reweighting the loss, outperforming random and heuristic baselines on ordering-sensitive tasks.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2606.00241","ref_index":233,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"InfoAtlas: A Foundation Model for Zero-Shot Statistical Dependence Estimate","primary_cat":"cs.LG","submitted_at":"2026-05-29T18:16:51+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"InfoAtlas is a pretrained neural model for zero-shot mutual information estimation that matches state-of-the-art accuracy with 100x speedup and handles varying dimensions via a single model.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.28267","ref_index":2,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"Parameter-Efficient Generative Modeling with Controlled Vector Fields","primary_cat":"cs.LG","submitted_at":"2026-05-27T10:14:29+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"Presents a controlled vector field framework for continuous generative modeling where velocity is formed from fixed bracket-generating fields modulated by scalar controls, with an expressivity principle under controllability assumptions.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.27064","ref_index":14,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"Flow-Based Global Proposals for Monte Carlo Sampling in SU(2) Lattice Gauge Theory","primary_cat":"hep-lat","submitted_at":"2026-05-26T14:15:26+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"A coupling-flow global proposal for Monte Carlo sampling in 2D pure SU(2) lattice gauge theory is shown to be formally valid and to reproduce the target ensemble in proof-of-principle tests, with modest hybrid gains but no clear outperformance over local baselines.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.25852","ref_index":4,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"A Post-Processing Conformal Prediction Approach for Conditional Coverage via Pivotal Scores","primary_cat":"stat.ME","submitted_at":"2026-05-25T13:44:25+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"PIT-CP post-processes nonconformity scores via one-dimensional conditional density estimation to produce approximately pivotal scores, achieving approximate conditional coverage in conformal prediction for i.i.d. data.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.23114","ref_index":21,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"Increasing the Precision of Surrogate Models for Weak Lensing Mass Maps with Flow Matching","primary_cat":"astro-ph.CO","submitted_at":"2026-05-22T00:20:49+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"A flow matching generative model produces weak lensing mass maps with fidelity improved to below 1% and 5% on basic and higher-order statistics relative to GAN benchmarks.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.22489","ref_index":293,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"Machine Learning Techniques for Astrophysics and Cosmology: Lyman-$\\alpha$ forest","primary_cat":"astro-ph.CO","submitted_at":"2026-05-21T13:43:46+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":2.0,"formal_verification":"none","one_line_summary":"Review of machine learning applications for analyzing Lyman-alpha forest observations to probe cosmology, reionization, and dark matter.","context_count":1,"top_context_role":"background","top_context_polarity":"background","context_text":"The blue arrow denotes the training stage, during which the emulator learns a bijective mapping between summary statistics measured from a suite of hydrodynamical simulations and an eight-dimensional normal distribu- tion. The mapping is conditioned on the cosmology and ionization and thermal state of the IGM in the simulations, and is implemented through affine coupling blocks (ACBs [293]). The green arrow indicates the emulation stage, in which the inverse mapping is applied to random samples drawn from the base distribution to generate predictions for Lyman-αforest clustering. Credit: Figure 3 from [289]. 1.6.3 Simulation-based inference Simulation-based inference (SBI [294-296]) provides a flexible alternative to tra- ditional likelihood-based approaches in Lyman-αforest analyses."},{"citing_arxiv_id":"2605.21957","ref_index":5,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"Bounding-Box Trajectories Matter for Video Anomaly Detection","primary_cat":"cs.CV","submitted_at":"2026-05-21T03:44:06+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"TrajVAD shows that bounding-box trajectories modeled via normalizing flows can serve as a primary cue for video anomaly detection, with the trajectory-only variant achieving 87.7% AP on ShanghaiTech and best results on MSAD.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.17742","ref_index":16,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"UST-Hand: An Uncertainty-aware Spatiotemporal Point Cloud Interaction Network for 3D Self-supervised Hand Pose Estimation","primary_cat":"cs.CV","submitted_at":"2026-05-18T01:47:13+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"UST-Hand is a self-supervised 3D hand pose estimation method using conditional normalizing flows for uncertainty-aware hypothesis sampling and probabilistic point cloud interactions to achieve up to 37.8% better MPVPE than prior self-supervised approaches on three datasets.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.16570","ref_index":259,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"A Cubing Strategy for Identifying Stable Hyperparameter Regions for Uncertainty Quantification in Spatial Deep Learning","primary_cat":"stat.CO","submitted_at":"2026-05-15T19:18:39+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"A recursive cubing framework identifies stable hyperparameter regions for MC dropout uncertainty quantification in spatial deep learning and produces competitive or superior predictive intervals versus a statistical baseline on simulations and land-surface temperature data.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.15579","ref_index":19,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"TVRN: Invertible Neural Networks for Compression-Aware Temporal Video Rescaling","primary_cat":"eess.IV","submitted_at":"2026-05-15T03:44:14+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"TVRN combines invertible wavelet-based networks with a surrogate gradient approximator and compression-aware asymmetric design to improve frame-rate rescaling quality under real codecs.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.12183","ref_index":7,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"DriftXpress: Faster Drifting Models via Projected RKHS Fields","primary_cat":"cs.LG","submitted_at":"2026-05-12T14:26:31+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"DriftXpress approximates drifting kernels via projected RKHS fields to lower training cost of one-step generative models while matching original FID scores.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.11199","ref_index":29,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"Operator Spectroscopy of Trained Lattice Samplers","primary_cat":"hep-lat","submitted_at":"2026-05-11T20:06:15+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"Operator projections of trained sampler functions in 2D phi^4 lattice theory decompose residuals into zero-mode Binder and finite-k correlator components, distinguishing flow-matching, diffusion, and normalizing-flow models.","context_count":1,"top_context_role":"background","top_context_polarity":"background","context_text":"respect to a baselineBis O⊥(t, ϕ) =O(ϕ)−Proj ρmatch(t) B O(ϕ).(27) The projection is calibrated once on ρmatch(t) and is not recomputed during rollout. This isolates the candidate's independent direction relative to the baseline. Raw monomials M, M 3, M5, . . . are highly collinear on ρmatch(t). We define per-tGram-Schmidt polynomials P1(M) =M,(28) P3(M;t) =M 3 −a 31(t)M,(29) P5(M;t) =M 5 −a 53(t)P3(M;t)−a 51(t)P1(M),(30) with ⟨PiPj⟩ρmatch(t) = 0 for i̸ = j. Adding P5(M; t) to BL4 selectively reduces the zero-mode residual and moves U4 toward the teacher/HMC value without changing G(2) orS(k min). The free FM resolvent predicts sensitivity to soft Fourier modes. On a finite lattice define ϕ|n|2=q =F −1 h 1n2x+n2y=q ˆϕn"},{"citing_arxiv_id":"2605.08561","ref_index":4,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"CONTRA: Conformal Prediction Region via Normalizing Flow Transformation","primary_cat":"stat.ML","submitted_at":"2026-05-08T23:43:14+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"CONTRA generates sharp multi-dimensional conformal prediction regions by defining nonconformity scores as distances from the center in the latent space of a normalizing flow.","context_count":1,"top_context_role":"method","top_context_polarity":"use_method","context_text":"Given a generative model for estimating the conditional density, PCP defines the non-conformity score of a data point (xi,y i)as si = min 1≤k≤K ∥yi − ˆyk i ∥, where{ ˆyi},K k=1 is a sample of the output generated from the estimated conditional density. Then the prediction region atx n+1 is obtained by generating a sample{ ˆyk n+1}K k=1 and form bCPCP(xn+1) = K[ k=1 {y:∥y− ˆyk n+1∥2 ≤s 1−α}, wheres 1−α is the⌈(1−α)(n 2 + 1)⌉-th smallest member of{s i, i∈I 2}. Note that each PCP prediction region is the union ofKballs. They have flexible, but often irregular, disconnected shapes, and are sensitive to the choice ofKandα. This brings challenge to interpreting the regions. Next, the ST-DQR method was proposed by Feldman et al. (2023). It learns an r-dimensional latent representation of the output, e.g., using the conditional variational auto-encoder (CV AE). The latent variable is encouraged to follow a unimodal distribution, so that methods like the directional quantile regression (DQR) are applicable to form convex probability regions for it. Samples are generated in the output space corresponding to points in the latent region. Calibration and the final prediction region are created similarly to the PCP. ST-DQR and our CONTRA are similar in that they both depend on latent representations, but differ substantially in the latent representation. CONTRA uses bijection and are often able to learn the latent variable to follow the Gaussian reference distribution rather closely. In contrast, the latent variable in ST-DQR is typically of lower dimension than the output, and its distribution is only coarsely similar to a reference. And this is why additional steps like DQR are needed to form latent probability regions in ST-DQR, but our CONTRA can directly use HDR of the reference Gaussian. When calibrating the regions, ST-DQR (and PCP) had to introduce yet another step: union balls around generated samples, and calibrate the common radius of the balls. Whereas CONTR"},{"citing_arxiv_id":"2605.08078","ref_index":10,"ref_count":2,"confidence":0.98,"is_internal_anchor":true,"paper_title":"Normalizing Trajectory Models","primary_cat":"cs.CV","submitted_at":"2026-05-08T17:57:14+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"NTM models each generative reverse step as a conditional normalizing flow with a hybrid shallow-deep architecture, enabling exact-likelihood training and strong four-step sampling performance on text-to-image tasks.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.08029","ref_index":8,"ref_count":1,"confidence":0.9,"is_internal_anchor":true,"paper_title":"STARFlow2: Bridging Language Models and Normalizing Flows for Unified Multimodal Generation","primary_cat":"cs.CV","submitted_at":"2026-05-08T17:14:43+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"STARFlow2 presents an autoregressive flow-based architecture for unified multimodal text-image generation by interleaving a VLM stream with a TarFlow stream via residual skips and a unified latent space.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.07546","ref_index":14,"ref_count":1,"confidence":0.9,"is_internal_anchor":true,"paper_title":"On the Invariance and Generality of Neural Scaling Laws","primary_cat":"cs.LG","submitted_at":"2026-05-08T10:21:09+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"Neural scaling laws are invariant under bijective data transformations and change predictably with information resolution ρ under non-bijective transformations, enabling cross-domain transport of fitted exponents.","context_count":1,"top_context_role":"method","top_context_polarity":"use_method","context_text":"ImageNet-1K [13] and ResNets [24] on CIFAR-100 [31]. Speech: Wav2Vec2 [4] on LibriSpeech [39]. Clinical time-series: MIMIC-IV [29] ICU vitals and labs. Each scaling-law fit uses 4-6 model capacities and 5-6 dataset sizes. We probe two bijective transformation families: linear (full- rank random matrix on token IDs) and nonlinear (affine coupling layers from the normalizing flows literature [14]), and two non-bijective families: quantization (Eq. (15)) and low-rank projection (Eq. (17)) of embeddings at multiple intensity levels. Full details are deferred to Appendix F. NSLs Under Various Transformations.Figure 1 (a)-(c) shows that NSLs remain largely invari- ant under bijective transformations, with linear and nonlinear re-encodings tracing nearly identical"},{"citing_arxiv_id":"2605.07319","ref_index":13,"ref_count":1,"confidence":0.9,"is_internal_anchor":true,"paper_title":"Generative Modeling with Flux Matching","primary_cat":"cs.LG","submitted_at":"2026-05-08T06:28:23+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":8.0,"formal_verification":"none","one_line_summary":"Flux Matching generalizes score-based generative modeling by using a weaker objective that admits infinitely many non-conservative vector fields with the data as stationary distribution, enabling new design choices beyond traditional score matching.","context_count":1,"top_context_role":"background","top_context_polarity":"background","context_text":"interpretable fields like RNA velocity, and structured generative dynamics. 2 Preliminaries Let pdata denote an unknown data distribution on Rd, observed only through samples {xi}n i=1 ∼ pdata. A key goal of generative modeling is to learn a representation of pdata that allows us to generate new samples from this distribution. Existing approaches do this by either modeling the density itself [13, 47], an unnormalized density [15, 39, 22, 24], or the closely related score function [31, 59, 25]. 2.1 The (Stein) score function∇logp data(x) Learning.If the score was directly available, we could fit a vector field fθ :R d →R d by minimizing the Fisher divergence: J(θ) =E x∼pdata \u0002 ∥fθ(x)− ∇logp data(x)∥2\u0003 .(1) However, the score ∇logp data(x) is typically inaccessible because pdata itself is unknown."},{"citing_arxiv_id":"2605.07100","ref_index":11,"ref_count":1,"confidence":0.9,"is_internal_anchor":true,"paper_title":"TRACE: Transport Alignment Conformal Prediction via Diffusion and Flow Matching Models","primary_cat":"stat.ML","submitted_at":"2026-05-08T01:28:57+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"TRACE creates valid conformal prediction sets for complex generative models by scoring outputs via averaged denoising or velocity errors along stochastic transport paths instead of likelihoods.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.06980","ref_index":5,"ref_count":1,"confidence":0.9,"is_internal_anchor":true,"paper_title":"Accelerating the Simulation of Ordinary Differential Equations Through Physics-Preserving Neural Networks","primary_cat":"math.NA","submitted_at":"2026-05-07T21:58:47+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"A neural network maps ODE states to a slow-evolving latent space with dynamics derived from the original equations via the chain rule, enabling accelerated simulations with fewer function calls.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.06905","ref_index":33,"ref_count":1,"confidence":0.9,"is_internal_anchor":true,"paper_title":"Conservative Flows: A New Paradigm of Generative Models","primary_cat":"cs.LG","submitted_at":"2026-05-07T20:06:08+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"Conservative flows generate by running probability-preserving stochastic dynamics initialized at data points rather than noise, using corrected Langevin or predictor-corrector mechanisms on top of any pretrained flow model and showing gains on Swiss-roll, ImageNet-256 and Oxford Flowers-102.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.06606","ref_index":61,"ref_count":2,"confidence":0.98,"is_internal_anchor":true,"paper_title":"TMDs in the Lens of Generative AI: A Pixel-Based Approach to Partonic Imaging","primary_cat":"hep-ph","submitted_at":"2026-05-07T17:25:41+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"A nonparametric pixel-based Bayesian method integrates TMD evolution with generative AI sampling and SVD to extract parton distributions and identify unconstrained null components from multi-scale observables.","context_count":1,"top_context_role":"method","top_context_polarity":"use_method","context_text":"1: (i)density approximationviaaniterativeNFtrainingloop, (ii)exact sampling through the MH chain, and (iii)final inferenceof the distributions. In the first phase, a generative NF model is trained to approximate the target pos- terior through an iterativetraining loop, as depicted in Fig. 1. The architecture consists of 12 masked affine coupling layers [61], interleaved with ActNorm modules and random permutations. This configuration is specifically designed to capture the complex, nonlocal correlations among the 50 nodes of thebT grid, effectively mapping the high-dimensional parameter space. For each iteration, the NF generates a batch of samples˜f, along with their associated log-probabilitieslnqNF( ˜f)."},{"citing_arxiv_id":"2605.06479","ref_index":289,"ref_count":1,"confidence":0.9,"is_internal_anchor":true,"paper_title":"Risk-Controlled Post-Processing of Decision Policies","primary_cat":"stat.ML","submitted_at":"2026-05-07T16:03:24+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"Risk-controlled post-processing yields a threshold-structured policy that follows the baseline except where an oracle fallback sharply reduces conditional violation risk, achieving O(log n/n) expected excess risk in i.i.d. settings and exact risk control under exchangeability.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.05520","ref_index":51,"ref_count":1,"confidence":0.9,"is_internal_anchor":true,"paper_title":"Bayesian Rain Field Reconstruction using Commercial Microwave Links and Diffusion Model Priors","primary_cat":"cs.LG","submitted_at":"2026-05-06T23:36:46+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":"2605.01868","ref_index":26,"ref_count":1,"confidence":0.9,"is_internal_anchor":true,"paper_title":"Robust Conditional Conformal Prediction via Branched Normalizing Flow","primary_cat":"cs.LG","submitted_at":"2026-05-03T13:29:26+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"Branched Normalizing Flow improves conditional coverage robustness of conformal prediction under distribution shift by normalizing test inputs to the calibration distribution and mapping prediction sets back.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.01134","ref_index":136,"ref_count":1,"confidence":0.9,"is_internal_anchor":true,"paper_title":"To Use AI as Dice of Possibilities with Timing Computation","primary_cat":"cs.AI","submitted_at":"2026-05-01T22:25:29+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":"2604.25732","ref_index":41,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"Personalized Multi-Interest Modeling for Cross-Domain Recommendation to Cold-Start Users","primary_cat":"cs.IR","submitted_at":"2026-04-28T15:01:27+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"NF-NPCDR enhances neural processes with normalizing flows to model personalized multi-interest preferences and uses a preference pool plus adaptive decoder to improve cross-domain recommendations for cold-start users.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2604.24330","ref_index":58,"ref_count":1,"confidence":0.9,"is_internal_anchor":true,"paper_title":"Pre-localization of Massive Black Hole Binaries in the Millihertz Band","primary_cat":"gr-qc","submitted_at":"2026-04-27T11:21:04+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"A neural spline flow pipeline performs amortized inference on millihertz MBHB signals, delivering ~20 deg² pre-merger sky localizations in ~1 minute while matching PTMCMC sky modes and parameter uncertainties.","context_count":1,"top_context_role":"method","top_context_polarity":"use_method","context_text":"Efficient computation of the last term in Eq. (22), the Jacobian determinant, ensures that volume changes in- duced by the transformation are properly accounted for, which is crucial for maintaining a valid probability den- sity. In this work, we employ a coupling NSF [57], in which each invertible layer is implemented as a coupling transformation [58] parameterized by rational quadratic splines (RQ-splines). Let aK-dimensional input vector be denoted asx∈R K. For thenth coupling layer, let the input be split as fn−1 = (xA,x B)∈R k ×R K−k , and let the output bef n = (yA,y B). The coupling map fn is defined by yA =x A, yB,i =g n,i(xB,i; Θn,i(xA)), i= 1, . . . , K−k, (23) 8 where the subscriptidenotes thei-th component of the"},{"citing_arxiv_id":"2604.24322","ref_index":25,"ref_count":1,"confidence":0.9,"is_internal_anchor":true,"paper_title":"Generative Design of a Gas Turbine Combustor Using Invertible Neural Networks","primary_cat":"cs.AI","submitted_at":"2026-04-27T11:14:06+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"Invertible Neural Networks are used to generate gas turbine combustor designs that meet specified performance criteria from a training database of parameterized designs and simulations.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2604.20041","ref_index":2,"ref_count":1,"confidence":0.9,"is_internal_anchor":true,"paper_title":"Normalizing Flows with Iterative Denoising","primary_cat":"cs.CV","submitted_at":"2026-04-21T22:52:26+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"iTARFlow augments normalizing flows with diffusion-style iterative denoising during sampling while preserving end-to-end likelihood training, reaching competitive results on ImageNet 64/128/256.","context_count":1,"top_context_role":"baseline","top_context_polarity":"baseline","context_text":"quality samples, while ten steps are adequate to reach opti- mal quantitative performance, measured by Fr'echet Incep- tion Distance (FID) (Heusel et al., 2017). During the de- noising phase, we do not apply CFG, as it does not provide Table 2: Overview of used network architectures Model num. Attention Layer @ each Transformer block patch size Channel Size Small (S) Model [2,2,2,12] Resolution/32 1280 Big (B) Model [4,4,4,24] Resolution/32 1280 Large (L) Model [4,4,4,24] Resolution/32 1600 Extra Large (XL) Model [4,4,4,24] Resolution/32 2176 Table 3: Comparison of models on ImageNet-64. Model FID↓# Param. Diffusion Models EDM-SDE (511 NFE) (Karras et al., 2022) 1.55 300M EDM-ODE (79 NFE) (Karras et al., 2022) 2.36 300M"},{"citing_arxiv_id":"2604.19087","ref_index":2,"ref_count":1,"confidence":0.9,"is_internal_anchor":true,"paper_title":"OLLM: Options-based Large Language Models","primary_cat":"cs.AI","submitted_at":"2026-04-21T04:59:37+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"OLLM models next-token generation as a latent-indexed set of options, enabling up to 70% math reasoning correctness versus 51% baselines and structure-based alignment via a compact latent policy.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2604.18339","ref_index":25,"ref_count":1,"confidence":0.9,"is_internal_anchor":true,"paper_title":"Probing the 3D Structures of Supernovae through IR Signatures of CO and SiO","primary_cat":"astro-ph.HE","submitted_at":"2026-04-20T14:39:40+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"MOFAT applied to SN2024ggi shows CO triggering inner SiO formation with a receding edge, order-of-magnitude mass drop, clumping signatures, and no dust formation.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2604.17266","ref_index":6,"ref_count":1,"confidence":0.9,"is_internal_anchor":true,"paper_title":"Scalable DDPM-Polycube: An Extended Diffusion-Based Method for Hexahedral Mesh and Volumetric Spline Construction","primary_cat":"cs.CE","submitted_at":"2026-04-19T05:41:19+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":3.0,"formal_verification":"none","one_line_summary":"Scalable DDPM-Polycube adds a blind-hole cube primitive, enlarges the grid to 3D, and introduces genus-guided hierarchical verification to improve diffusion-based polycube generation for complex geometries.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2604.11636","ref_index":8,"ref_count":1,"confidence":0.9,"is_internal_anchor":true,"paper_title":"MorphoFlow: Sparse-Supervised Generative Shape Modeling with Adaptive Latent Relevance","primary_cat":"cs.CV","submitted_at":"2026-04-13T15:45:59+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"MorphoFlow learns compact probabilistic 3D shape representations from sparse annotations using neural implicits, autodecoders, autoregressive flows, and adaptive sparsity priors on latent dimensions.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2604.09769","ref_index":11,"ref_count":1,"confidence":0.9,"is_internal_anchor":true,"paper_title":"Differentiable free energy surface: a variational approach to directly observing rare events using generative deep-learning models","primary_cat":"physics.comp-ph","submitted_at":"2026-04-10T18:00:08+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"VaFES constructs a latent space from reversible collective variables and variationally optimizes a tractable-density generative model to produce a continuous free energy surface from which rare events are directly sampled.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.04060","ref_index":7,"ref_count":1,"confidence":0.9,"is_internal_anchor":true,"paper_title":"Lookahead Drifting Model","primary_cat":"cs.LG","submitted_at":"2026-04-10T11:12:21+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"The lookahead drifting model improves upon the drifting model by sequentially computing multiple drifting terms that incorporate higher-order gradient information, leading to better performance on toy examples and CIFAR10.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2604.06348","ref_index":33,"ref_count":1,"confidence":0.9,"is_internal_anchor":true,"paper_title":"Dartmouth Stellar Evolution Emulator (DSEE) 1: Generative Stellar Evolution Model Database","primary_cat":"astro-ph.SR","submitted_at":"2026-04-07T18:27:53+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"DSEE is a flow-based emulator that generates stellar evolution tracks and isochrones as probabilistic outputs from a single model trained on millions of simulations, enabling fast interpolation and uncertainty-aware analyses.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2604.05303","ref_index":33,"ref_count":1,"confidence":0.9,"is_internal_anchor":true,"paper_title":"Jeffreys Flow: Robust Boltzmann Generators for Rare Event Sampling via Parallel Tempering Distillation","primary_cat":"cs.LG","submitted_at":"2026-04-07T01:17:53+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"Jeffreys Flow distills Parallel Tempering trajectories via Jeffreys divergence to produce robust Boltzmann generators that suppress mode collapse and correct sampling inaccuracies for rare event sampling.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2604.08582","ref_index":47,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"Multivariate Time Series Anomaly Detection via Dual-Branch Reconstruction and Autoregressive Flow-based Residual Density Estimation","primary_cat":"cs.LG","submitted_at":"2026-03-29T15:18:14+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"DBR-AF decouples cross-variable correlations in reconstruction and applies autoregressive flows to model residual densities for improved anomaly detection in multivariate time series.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2603.26357","ref_index":17,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"MPDiT: Multi-Patch Global-to-Local Transformer Architecture For Efficient Flow Matching and Diffusion Model","primary_cat":"cs.CV","submitted_at":"2026-03-27T12:30:10+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"MPDiT uses a hierarchical multi-patch design in transformers to lower computation in diffusion models by handling coarse global features first then fine local details, plus faster-converging embeddings.","context_count":1,"top_context_role":"background","top_context_polarity":"background","context_text":"ing convergence. Extensive experiments on the ImageNet dataset demonstrate the effectiveness of our architectural choices. Code is released athttps://github.com/ quandao10/MPDiT 1. Introduction Diffusion models [16, 27, 37, 58] have emerged as a lead- ing class of generative models, surpassing generative adver- sarial networks [20], normalizing flows [17, 33, 80], and autoregressive models [45, 61, 64] in many vision tasks. Compared to GANs [20], diffusion models [27] are gen- erally easier to train and avoid issues such as instability and mode collapse. In 2D image generation, diffusion-based approaches have demonstrated strong performance in text- to-image synthesis [52], enabling downstream applications"},{"citing_arxiv_id":"2603.14135","ref_index":30,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"Conditional flow matching for physics-constrained inverse problems with finite training data","primary_cat":"stat.ML","submitted_at":"2026-03-14T21:43:48+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"Conditional flow matching learns a velocity field to sample from measurement-conditioned posteriors in physics inverse problems, with early stopping to prevent variance collapse and selective memorization under finite training data.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2602.18560","ref_index":64,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"Inferring the population properties of galactic binaries from LISA's stochastic foreground","primary_cat":"astro-ph.HE","submitted_at":"2026-02-20T19:00:07+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"A neural posterior estimator trained on simulated LISA foreground spectra recovers galactic binary population parameters, including total number, with good accuracy in validation tests.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2602.09580","ref_index":13,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"SERNF: Sample-Efficient Real-World Dexterous Policy Fine-Tuning via Action-Chunked Critics and Normalizing Flows","primary_cat":"cs.RO","submitted_at":"2026-02-10T09:28:20+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":"2602.07633","ref_index":14,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"Flow-Based Conformal Predictive Distributions","primary_cat":"stat.ML","submitted_at":"2026-02-07T17:26:50+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"Differentiable nonconformity scores induce flows that sample conformal prediction set boundaries, and mixing flows across levels produces conformal predictive distributions whose quantiles match the sets.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2601.15739","ref_index":9,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"Breaking the Resolution Barrier: Arbitrary-resolution Deep Image Steganography Framework","primary_cat":"cs.CV","submitted_at":"2026-01-22T08:07:10+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"ARDIS enables arbitrary-resolution deep image steganography via frequency decoupling in hiding and latent-guided implicit reconstruction for blind recovery.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2512.18928","ref_index":31,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"The Ensemble Schr{\\\"o}dinger Bridge filter for Nonlinear Data Assimilation","primary_cat":"cs.LG","submitted_at":"2025-12-22T00:06:49+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"The Ensemble Schrödinger Bridge filter adds a diffusion-based analysis step to ensemble prediction, enabling effective nonlinear data assimilation without structural model error or training.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2511.00956","ref_index":35,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"RefTon: Reference person shot assist virtual Try-on","primary_cat":"cs.CV","submitted_at":"2025-11-02T14:32:31+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"RefTon is a flux-based virtual try-on method that uses unpaired reference images of the target garment on different people to guide texture and detail preservation in a streamlined person-to-person pipeline without body parsing or masks.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2510.06508","ref_index":34,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"Application of deep neural networks for computing the renormalization group flow of the two-dimensional phi^4 field theory","primary_cat":"cond-mat.dis-nn","submitted_at":"2025-10-07T22:54:55+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"RGFlow uses flow-based neural networks to learn bijective real-space RG transformations for the 2D phi^4 theory, identifying a Wilson-Fisher-like critical point and estimating the correlation length exponent.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null}],"limit":50,"offset":0}