{"total":13,"items":[{"citing_arxiv_id":"2606.01634","ref_index":11,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"E4GEN: Event-level Explainable Extreme-Enhanced Time-series Generation","primary_cat":"cs.LG","submitted_at":"2026-06-01T03:35:32+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"E4GEN is an explainable diffusion model using E-Activator, E-Predictor, and E-Control for extreme-event-aware time-series generation evaluated on six datasets.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2606.09861","ref_index":11,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Time Series as Language: A Universal Tokenizer for General-Purpose Time Series Foundation Models","primary_cat":"cs.LG","submitted_at":"2026-05-31T16:04:11+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"UniTok tokenizes time series for an off-the-shelf LLM foundation model that unifies forecasting, generation, and classification through next-token prediction and training-free inference.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2606.00708","ref_index":49,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"MOSAIC: Modular Orchestration for Structured Agentic Intelligence and Composition","primary_cat":"cs.AI","submitted_at":"2026-05-30T12:31:13+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"MOSAIC structures LLM-based model selection via memory-grounded blueprints and failure-aware RL, reporting gains in performance and traceability on financial time-series tasks over AutoML and agent baselines.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2606.07605","ref_index":2,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"SRT: Super-Resolution for Time Series via Disentangled Rectified Flow","primary_cat":"cs.LG","submitted_at":"2026-05-29T10:12:45+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"SRT decomposes low-resolution time series into trend and seasonal components, aligns them via implicit neural representations, and uses cross-resolution attention within a disentangled rectified flow to generate high-resolution outputs, with a scaled SRT-large variant for zero-shot use.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.28507","ref_index":18,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Universal Time Series Generation with Neural Controlled Differential Equations","primary_cat":"cs.LG","submitted_at":"2026-05-27T14:10:41+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"Proves SLiCEs are universal time-series generators approximating path laws in W_∞ and proposes G-SLiCEs for path-space flow matching with benefits on irregular grids.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2606.07569","ref_index":20,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"TriHead-GAN: A Generative Adversarial Network with Triple-Head Discriminator for Carbon Emission Time Series Generation","primary_cat":"cs.LG","submitted_at":"2026-05-26T02:48:56+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":4.0,"formal_verification":"none","one_line_summary":"TriHead-GAN is a GAN framework whose triple-head discriminator supervises distributional authenticity, cross-variable dependency via regression, and temporal smoothness via adjacent-difference prediction for carbon emission time series.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.28867","ref_index":31,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"PrismFlow: Residual Dynamics for Flow Matching in Time-Series Generation","primary_cat":"cs.LG","submitted_at":"2026-05-22T07:10:20+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"PrismFlow augments flow matching with residual dynamical experts and a winner-take-all objective to reduce spectral distortion and improve mode coverage in time-series generation.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.17804","ref_index":13,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"GenTS: A Comprehensive Benchmark Library for Generative Time Series Models","primary_cat":"cs.LG","submitted_at":"2026-05-18T03:27:54+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"GenTS is a modular benchmark library providing unified data pipelines, generative models, and evaluation metrics for time series synthesis, forecasting, and imputation, with open-source code and initial benchmarking experiments.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.05736","ref_index":9,"ref_count":2,"confidence":0.9,"is_internal_anchor":false,"paper_title":"SDFlow: Similarity-Driven Flow Matching for Time Series Generation","primary_cat":"cs.AI","submitted_at":"2026-05-07T06:28:18+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"SDFlow learns a global transport map via similarity-driven flow matching in VQ latent space, using low-rank manifold decomposition and a categorical posterior to handle discreteness, yielding SOTA long-horizon performance and inference speedups.","context_count":1,"top_context_role":"baseline","top_context_polarity":"baseline","context_text":",xℓ)∈R ℓ×d denote a multivariate time series of length ℓwithdobserved dimensions. The unconditional generation problem is defined as: Input:Z 0 ∼π 0;Output: ˆX1:ℓ =G(Z 0)∈R ℓ×d,(1) where Z0 ∈R ℓ×d, π0 =N(0,I) is a tractable prior, and G maps samples from the prior to the target data distribution. The generative model can be implemented using GANs [29], V AEs [9], or diffusion models [30], which effectively capture complex temporal dependencies. During training, G is optimized to minimize discrepancy between generated ˆX1:ℓ and real data X1:ℓ, ensuring high fidelity and diversity in generated sequences. Similarity-Driven Vector Quantization.Following [ 7], we employ a VQ-V AE tokenizer that maps continuous time series to discrete token sequences."},{"citing_arxiv_id":"2603.08032","ref_index":4,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"GCGNet: Graph-Consistent Generative Network for Time Series Forecasting with Exogenous Variables","primary_cat":"cs.LG","submitted_at":"2026-03-09T07:11:01+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"GCGNet uses a variational generator, graph structure aligner, and graph refiner to jointly capture temporal and channel correlations in time series forecasting with exogenous variables, outperforming baselines on 12 real-world datasets.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2602.08299","ref_index":13,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Cyclic Adaptive Private Synthesis for Sharing Real-World Data in Education","primary_cat":"cs.CY","submitted_at":"2026-02-09T06:09:18+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"CAPS provides an iterative differentially private synthesis method that outperforms one-shot baselines on authentic educational real-world data.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2505.20662","ref_index":1,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"AutoReproduce: Automatic AI Experiment Reproduction with Paper Lineage","primary_cat":"cs.AI","submitted_at":"2025-05-27T03:15:21+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"AutoReproduce is a multi-agent system using paper lineage to autonomously reproduce AI experiment code, with a new benchmark showing improvements over baselines in fidelity and execution.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2505.14202","ref_index":22,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"MSDformer: Multi-scale Discrete Transformer For Time Series Generation","primary_cat":"cs.LG","submitted_at":"2025-05-20T11:01:36+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"MSDformer introduces a multi-scale discrete transformer that tokenizes time series at multiple scales and models them autoregressively in discrete space, claiming superior performance over prior DTM methods with rate-distortion theoretical support.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null}],"limit":50,"offset":0}