Research map / Time series / Long-term forecasting
CoRA: Boosting Time Series Foundation Models for Multivariate Forecasting through Correlation-aware Adapter
Hanyin Cheng, Xingjian Wu, Yang Shu, Zhongwen Rao, Lujia Pan, Bin Yang, Chenjuan Guo
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 CoRA adapter during fine-tuning | GPT4TS · ETTh1 (5% few-shot) | MSE: 0.468 -> 0.456 | Improved |
| add CoRA adapter during fine-tuning | UniTime · ETTh1 (5% few-shot) | MSE: 0.739 -> 0.714 | Improved |
| add CoRA adapter during fine-tuning | TTM · Electricity (5% few-shot) | MSE: 0.179 -> 0.177 | Improved |
| add CoRA adapter during fine-tuning | 6 TSFMs · 10 datasets (5% few-shot) | MSE | Improved |
| add H-PCorr contrastive learning alone (ablation) | GPT4TS + naive modules · ETTm2 | MSE: 0.279 -> 0.277 | Improved |
| add Dynamic Correlation Estimation to HPCL (ablation) | GPT4TS · ETTm2 | MSE: 0.277 -> 0.273 | Improved |
| add Heterogeneous Division to HPCL (ablation) | GPT4TS · ETTm2 | MSE: 0.277 -> 0.274 | Improved |
| combine all three modules (ablation) | GPT4TS · Electricity | MSE: 0.206 -> 0.201 | Improved |
| add CoRA with only 3% training data | TTM · ETTm2 | MSE: 0.263 -> 0.261 | Improved |
| add CoRA with 3% training data | TTM · Weather | MSE: 0.237 -> 0.234 | Improved |
| add CoRA at 20% training data | CALF · ETTm2 | MSE: 0.261 -> 0.254 | Improved |
| add LIFT or C-LoRA plugins instead of CoRA | GPT4TS, UniTime, Timer · ETTm2, Weather, Electricity (5% few-shot) | MSE | Hurt |
Closest papers by meaning
- AdaPTS: Adapting Univariate Foundation Models to Probabilistic Multivariate Time Series Forecasting (ICML 2025 · similarity 0.89)
- ARM: Refining Multivariate Forecasting with Adaptive Temporal-Contextual Learning (ICLR 2024 · similarity 0.88)
- MSGNet: Learning Multi-Scale Inter-series Correlations for Multivariate Time Series Forecasting (AAAI 2024 · similarity 0.87)
- Revitalizing Multivariate Time Series Forecasting: Learnable Decomposition with Inter-Series Dependencies and Intra-Series Variations Modeling (ICML 2024 · similarity 0.87)
- Multivariate Probabilistic Time Series Forecasting with Correlated Errors (NeurIPS 2024 · similarity 0.87)
- TimePFN: Effective Multivariate Time Series Forecasting with Synthetic Data (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.
