Research map / Time series / Foundation models for time series
TimeOmni-VL: Unified Models for Time Series Understanding and Generation
Tong Guan, Sheng Pan, Johan Barthelemy, Zhao Li, Yujun Cai, Cesare Alippi, Ming Jin, Shirui Pan
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| replace the heatmap TS-image with the TS2I periodic-cycle layout (ablation) | TimeOmni-VL · GIFT-Eval forecasting and imputation | nMASE: heatmap variant yields nMASE worse than the Naive baseline on nearly all tasks | Improved |
| add generation CoT from the understanding model as context (ablation) | TimeOmni-VL · GIFT-Eval forecasting and imputation | nMASE: average 8.2% increase in nMASE without CoT | Improved |
| post-train Bagel-7B on TSUMM-Suite for forecasting | Bagel-7B · GIFT-Eval forecasting, short-term | nMASE: 16.303 -> 0.878 | Improved |
| post-train Bagel-7B on TSUMM-Suite for imputation | Bagel-7B · GIFT-Eval imputation, mask [0.1,0.2) | nMASE: 17.411 -> 0.713 | Improved |
| post-train Bagel on TS-image understanding QA3 | Bagel-7B · TSUMM-Suite understanding QA3 | score: 0.012 -> 0.931 | Improved |
| post-train Bagel on TS-image understanding QA5 | Bagel-7B · TSUMM-Suite understanding QA5 | score: 0 -> 0.667 | Improved |
| compare TimeOmni-VL with VisionTS++ retrained on the same 5k samples, forecasting average | TimeOmni-VL · GIFT-Eval forecasting average | nMASE: 0.900 -> 0.826 | Improved |
| compare TimeOmni-VL with VisionTS++ retrained on the same 5k samples, imputation average | TimeOmni-VL · GIFT-Eval imputation average | nMASE: 0.827 -> 0.810 | Improved |
| compare TimeOmni-VL with VisionTS++ (original), long-term forecasting | TimeOmni-VL · GIFT-Eval forecasting, long-term | nMASE: 0.690 -> 0.784 | Hurt |
| compare TimeOmni-VL with TimeOmni-1 on reasoning Task1 | TimeOmni-VL · TSR-Suite Task1 | accuracy: 87.7 -> 84.0 | Hurt |
| compare TimeOmni-VL with linear interpolation, imputation mask [0.4,0.5] | TimeOmni-VL · GIFT-Eval imputation, mask [0.4,0.5] | nMASE: 0.968 -> 0.927 | No clear effect |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 4 on most shared results.
- VisionTS++ — VisionTS: Visual Masked Autoencoders Are Free-Lunch Zero-Shot Time Series Forecasters (ICML 2025) · better on 3 of 6 shared results
- Moment-large — MOMENT: A Family of Open Time-series Foundation Models (ICML 2024) · better on 4 of 4 shared results
- ChatTime — ChatTime: A Unified Multimodal Time Series Foundation Model Bridging Numerical and Textual Data (AAAI 2025) · better on 3 of 3 shared results
- TimeOmni-1 — TimeOmni-1: Incentivizing Complex Reasoning with Time Series in Large Language Models (ICLR 2026) · better on 0 of 1 shared result
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Time-VLM: Exploring Multimodal Vision-Language Models for Augmented Time Series Forecasting (ICML 2025 · similarity 0.90)
- Time Series Generation Under Data Scarcity: A Unified Generative Modeling Approach (NeurIPS 2025 · similarity 0.88)
- Empowering Time Series Analysis with Large-Scale Multimodal Pretraining (NeurIPS 2026 · similarity 0.87)
- Utilizing Image Transforms and Diffusion Models for Generative Modeling of Short and Long Time Series (NeurIPS 2024 · similarity 0.87)
- Time Weaver: A Conditional Time Series Generation Model (ICML 2024 · similarity 0.87)
- TimeMixer++: A General Time Series Pattern Machine for Universal Predictive Analysis (ICLR 2025 · similarity 0.87)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
