Research map / Time series / Long-term forecasting
CometNet: Contextual Motif-guided Long-term Time Series Forecasting
Weixu Wang, Xiaobo Zhou, Xin Qiao, Lei Wang, Tie Qiu
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 |
|---|---|---|---|
| add contextual motif-guided forecasting (CometNet vs TimeMixer++) | CometNet · ETTh2 H=720 | MSE: 0.392 -> 0.228 | Improved |
| CometNet vs TimeMixer++ on Traffic | CometNet · Traffic avg | MSE: 0.416 -> 0.422 | Hurt |
| CometNet vs TimeMixer++ on Electricity | CometNet · Electricity avg | MSE: 0.165 -> 0.158 | Mixed |
| add motif components (all) vs none (ablation) | CometNet · ETTh1 H=720 | MSE: 0.544 -> 0.391 | Improved |
| add motif-driven gating (ablation) | CometNet · ETTh1 H=720 | MSE: 0.476 -> 0.391 | Improved |
| add position signal (ablation) | CometNet · ETTm2 H=720 | MSE: 0.262 -> 0.171 | Improved |
| add cross-scale redundancy filtering (ablation) | CometNet · ETTh2 H=720 | MSE: 0.492 -> 0.331 | Improved |
| add motif selection (ablation) | CometNet · ETTh1 H=720 | MSE: 0.430 -> 0.391 (MAE 0.470 -> 0.434) | Mixed |
| increase number of motifs K from 3 to 10 (ablation) | CometNet · ETT datasets | MSE | Improved |
| increase number of motifs K beyond 10 (ablation) | CometNet · ETT datasets | MSE | No clear effect |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- TimeMixer++ — TimeMixer: Decomposable Multiscale Mixing for Time Series Forecasting (ICLR 2024) · better on 7 of 8 shared results
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
- TStruct: Learning Shared Temporal Structures for Long-Term Time-Series Forecasting (NeurIPS 2026 · similarity 0.87)
- Sequence Complementor: Complementing Transformers for Time Series Forecasting with Learnable Sequences (AAAI 2025 · similarity 0.87)
- Time-LLM: Time Series Forecasting by Reprogramming Large Language Models (ICLR 2024 · similarity 0.87)
- TimeBase: The Power of Minimalism in Efficient Long-term Time Series Forecasting (ICML 2025 · similarity 0.86)
- Enhancing Foundation Models for Time Series Forecasting via Wavelet-based Tokenization (ICML 2025 · similarity 0.86)
- TimeCAP: Learning to Contextualize, Augment, and Predict Time Series Events with Large Language Model Agents (AAAI 2025 · similarity 0.86)
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.
