{"total":17,"items":[{"citing_arxiv_id":"2605.25166","ref_index":4,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"AME-TS: Anchored Mixture-of-Experts for Time Series Forecasting","primary_cat":"cs.LG","submitted_at":"2026-05-24T16:52:48+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"AME-TS is a structure-guided sparse MoE foundation model for time series that aligns expert routing with series-level temporal descriptors to achieve strong accuracy-efficiency tradeoffs on GIFT-Eval while improving specialization stability.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.26161","ref_index":2,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"TSFMAudit: Data Contamination Auditing in Forecasting Time Series Foundation Models","primary_cat":"cs.LG","submitted_at":"2026-05-24T14:59:12+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":8.0,"formal_verification":"none","one_line_summary":"TSFMAudit detects pretraining contamination in time series foundation models via probe adaptation dynamics (faster loss drop, smaller backbone shift), tested on 6 models and 187 datasets against 10 LLM-derived baselines.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.20268","ref_index":56,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Chronicle: A Multimodal Foundation Model for Joint Language and Time Series Understanding","primary_cat":"cs.LG","submitted_at":"2026-05-18T21:39:05+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"Chronicle is the first model jointly pretrained from scratch on text and time series in a unified transformer that matches a comparable language model on NLU tasks and sets new bars for time series classification and multimodal forecasting.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.15465","ref_index":2,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Toward World Modeling of Physiological Signals with Chaos-Theoretic Balancing and Latent Dynamics","primary_cat":"cs.LG","submitted_at":"2026-05-14T23:06:15+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"NormWear-2 encodes physiological signals and interventions into a shared latent space, models their joint evolution as a dynamical system, and uses chaos-theoretic balancing during pretraining to achieve superior multi-scale forecasting on diverse real-world datasets.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.13986","ref_index":55,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"TabPFN-3: Technical Report","primary_cat":"cs.LG","submitted_at":"2026-05-13T18:01:43+00:00","verdict":null,"verdict_confidence":null,"novelty_score":null,"formal_verification":null,"one_line_summary":null,"context_count":1,"top_context_role":"baseline","top_context_polarity":"baseline","context_text":"Looking at the win-rate results, TabPFN-TS-3's ranking drops to the 4th place, although we found these rates to be very sensitive to tiny differences on a few datasets. The strong performance of TabPFN-TS-3 is particularly noteworthy seeing that it is trained purely on synthetic data, while most other time-series models, including Chronos-2 [53], TiRex [54] and TimesFM-2.5 [55] are trained on real-world data. This property of our model prevents many issues from real-data pretraining: historical series are leaky and frequently recirculated across forecasting libraries (fev-bench flags 10% leakage in TimesFM-2.5 and 28% in Moirai-2.0; see Table 1), forecasting the future from historical pretraining is fundamentally out-of-distribution, and the supply of public real-world time-series"},{"citing_arxiv_id":"2605.08857","ref_index":51,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"RareCP: Regime-Aware Retrieval for Efficient Conformal Prediction","primary_cat":"cs.LG","submitted_at":"2026-05-09T10:12:07+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"RareCP improves interval efficiency for time series conformal prediction by retrieving and weighting regime-specific calibration examples while adapting to drift and maintaining coverage.","context_count":1,"top_context_role":"baseline","top_context_polarity":"baseline","context_text":"Torkkola, Andrew Gordon Wilson, Michael Bohlke-Schneider, and Yuyang Wang. Chronos: Learning the language of time series, 2024. URLhttps://arxiv.org/abs/2403.07815. 13 [50] Gerald Woo, Chenghao Liu, Akshat Kumar, Caiming Xiong, Silvio Savarese, and Doyen Sahoo. Unified training of universal time series forecasting transformers, 2024. URL https: //arxiv.org/abs/2402.02592. [51] Abhimanyu Das, Weihao Kong, Rajat Sen, and Yichen Zhou. A decoder-only foundation model for time-series forecasting, 2024. URLhttps://arxiv.org/abs/2310.10688. [52] Shi Bin Hoo, Samuel Müller, David Salinas, and Frank Hutter. From Tables to Time: Extending TabPFN-v2 to Time Series Forecasting.arXiv e-prints, art. arXiv:2501.02945, January 2025. doi: 10."},{"citing_arxiv_id":"2604.22328","ref_index":31,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"FETS Benchmark: Foundation Models Outperform Dataset-specific Machine Learning in Energy Time Series Forecasting","primary_cat":"cs.LG","submitted_at":"2026-04-24T08:00:50+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"Foundation models outperform dataset-specific machine learning in energy time series forecasting across 54 datasets in 9 categories.","context_count":1,"top_context_role":"baseline","top_context_polarity":"baseline","context_text":"differ regarding their underlying architectural design and modeling assumptions. Several approaches leverage transformer-based sequential models operating on patch embeddings of the time series, e.g., Chronos-2 [14], TimesFM [28], and Moirai2 [29]. Further state- of-the-art approaches are FlowState [30], leveraging state space models, as well as TiRex 5 [31] building upon an xLSTM-based architecture [32]. For tabular data, TabPFN2.5-TS [33, 34] employs a transformer-based prior-data fitted network. Further distinction between these state-of-the-art approaches lies in the forecasting modes they support, i.e., univariate forecasting predicting a single time series without additional information, covariate-informed"},{"citing_arxiv_id":"2604.20122","ref_index":4,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Adaptive Conformal Anomaly Detection with Time Series Foundation Models for Signal Monitoring","primary_cat":"cs.LG","submitted_at":"2026-04-22T02:39:46+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"A model-agnostic adaptive conformal anomaly detection approach uses weighted quantile bounds learned from past foundation model predictions to deliver interpretable p-value scores with stable calibration under shifts for time series monitoring.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2604.11529","ref_index":7,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"TempusBench: An Evaluation Framework for Time-Series Forecasting","primary_cat":"cs.LG","submitted_at":"2026-04-13T14:29:34+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"TempusBench is a new evaluation framework for time-series forecasting models that supplies fresh non-overlapping datasets, tasks beyond horizon and domain, consistent tuning across models, and visualization tools.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2604.16428","ref_index":3,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Non-Stationarity in the Embedding Space of Time Series Foundation Models","primary_cat":"cs.LG","submitted_at":"2026-04-06T05:04:53+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"Embedding spaces of time series foundation models make mean shifts, variance changes, and trends linearly detectable, but detection degrades smoothly with shift strength and shows model-specific failure modes.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2603.25777","ref_index":24,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Challenges and opportunities for AI to help deliver fusion energy","primary_cat":"physics.plasm-ph","submitted_at":"2026-03-26T13:15:37+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":2.0,"formal_verification":"none","one_line_summary":"AI offers opportunities to advance fusion energy R&D but requires responsible practices and expert collaborations to overcome its inherent challenges.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2603.22586","ref_index":4,"ref_count":2,"confidence":0.9,"is_internal_anchor":false,"paper_title":"A Foundation Model for Instruction-Conditioned In-Context Time Series Tasks","primary_cat":"cs.LG","submitted_at":"2026-03-23T21:24:41+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"iAmTime is a time-series foundation model that uses instruction-conditioned in-context learning from demonstrations to perform zero-shot adaptation on forecasting, imputation, classification, and related tasks.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2603.04791","ref_index":5,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Timer-S1: A Billion-Scale Time Series Foundation Model with Serial Scaling","primary_cat":"cs.AI","submitted_at":"2026-03-05T04:13:57+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"Timer-S1 is a released 8.3B-parameter MoE time series model that achieves state-of-the-art MASE and CRPS scores on GIFT-Eval using serial scaling and Serial-Token Prediction.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2602.08318","ref_index":4,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Is Flow Matching Just Trajectory Replay for Sequential Data?","primary_cat":"stat.ML","submitted_at":"2026-02-09T06:48:45+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"Flow matching on time series targets a closed-form nonparametric velocity field that is a similarity-weighted mixture of observed transition velocities, making neural models approximations to an ideal memory-augmented dynamical system sampler.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2512.20761","ref_index":4,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"TS-Arena -- A Live Forecast Pre-Registration Platform","primary_cat":"cs.LG","submitted_at":"2025-12-23T20:48:11+00:00","verdict":"CONDITIONAL","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"TS-Arena is a live pre-registration platform that evaluates time series forecasts on future data streams to eliminate information leakage.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2511.00266","ref_index":12,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"X-TRACK: Physics-Aware xLSTM for Realistic Vehicle Trajectory Prediction","primary_cat":"cs.LG","submitted_at":"2025-10-31T21:33:46+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"X-TRACK is the first xLSTM model with explicit kinematic constraints that generates realistic highway trajectories and outperforms baselines on highD while matching SOTA on NGSIM.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2509.25826","ref_index":3,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Kairos: Toward Adaptive and Parameter-Efficient Time Series Foundation Models","primary_cat":"cs.LG","submitted_at":"2025-09-30T06:02:26+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"Kairos is a parameter-efficient time series foundation model using dynamic patching tokenizer, mixture-of-size encoding, and spectral-conditioned positional embeddings to improve zero-shot forecasting on heterogeneous data.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null}],"limit":50,"offset":0}