Research map / Time series / Foundation models for time series
Unlocking the Value of Text: Event-Driven Reasoning and Multi-Level Alignment for Time Series Forecasting
Siyuan Wang, Peng Chen, Yihang Wang, Wanghui Qiu, Chenjuan Guo, Bin Yang, Yang Shu
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 event-driven reasoning and alignment to a time series model (ablation) | VoT · Energy / Social Good | MSE: 0.250/0.944 -> 0.222/0.804 | Improved |
| Add endogenous text alignment (ETA) (ablation) | VoT · Energy / Social Good | MSE: 0.241/0.840 -> 0.222/0.804 | Improved |
| Add historical in-context learning (HIC) (ablation) | VoT · Energy / Social Good | MSE: 0.238/0.845 -> 0.222/0.804 | Improved |
| Add the event-driven branch with HIC guidance (ablation) | VoT · Energy / Social Good | MSE: 0.232/0.841 -> 0.222/0.804 | Improved |
| Retrieve and summarize history rather than raw time series (full HIC vs time series retrieval) (ablation) | HIC · Climate | MSE: 1.137 -> 1.078 | Improved |
| Retrieve raw time series instead of nothing (ablation) | HIC · Social Good | MSE: 0.845 -> 0.853 | Hurt |
| Use exogenous text instead of random text (ablation) | VoT event branch · Energy | MSE: 27.5% lower MSE and 14.3% lower MAE | Improved |
| Use exogenous text instead of random text (ablation) | VoT event branch · Social Good | MSE: 12.8% MSE reduction and 16.0% MAE reduction | Improved |
| Use trend-seasonal decomposition in ETA (ablation) | ETA · Social Good | MSE: 0.892 -> 0.841 | Improved |
| Add TS-Text contrastive learning to ETA (ablation) | ETA · Social Good | MSE: 0.944 -> 0.841 | Improved |
| Use ETA instead of cross-attention for alignment (ablation) | VoT · Climate | MSE: 1.202 -> 1.092 | Improved |
| Use 6-parameter adaptive frequency fusion instead of MLP (ablation) | VoT · Energy | MSE: 0.685 -> 0.222 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.
- CALF — CALF: Aligning LLMs for Time Series Forecasting via Cross-modal Fine-Tuning (AAAI 2025) · better on 3 of 3 shared results
- iTransformer — iTransformer: Inverted Transformers Are Effective for Time Series Forecasting (ICLR 2024) · better on 3 of 3 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
- From News to Forecast: Integrating Event Analysis in LLM-Based Time Series Forecasting with Reflection (NeurIPS 2024 · similarity 0.89)
- Time-LLM: Time Series Forecasting by Reprogramming Large Language Models (ICLR 2024 · similarity 0.89)
- Context is Key: A Benchmark for Forecasting with Essential Textual Information (ICML 2025 · similarity 0.89)
- TimeXL: Explainable Multi-modal Time Series Prediction with LLM-in-the-Loop (NeurIPS 2025 · similarity 0.89)
- Rethinking Post-Training Recipes for Multimodal Time-Series Forecasting (NeurIPS 2026 · similarity 0.89)
- STEM-LTS: Integrating Semantic-Temporal Dynamics in LLM-driven Time Series Analysis (AAAI 2025 · similarity 0.89)
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.
