Research map / Time series / Long-term forecasting
HELIX: Hybrid Encoding with Learnable Identity and Cross-dimensional Synthesis for Time Series Imputation
Fengming Zhang, Wenjie Du, Huan Zhang, Ke Yu, Shen Qu
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 Feature Identity Embedding (ablation) | HELIX · BeijingAir Subseq-50% | MAE: 0.398 -> 0.166 | Improved |
| Add Feature Identity Embedding (ablation) | HELIX · ETT-h1 Point-50% | MAE: .241 -> .189 | Improved |
| Add Feature Identity Embedding (ablation) | HELIX · BeijingAir Block-50% | MAE: 0.223 -> 0.131 | Improved |
| Add double-helix hybrid encoding over serial Time-Feature-Time (ablation) | HELIX · BeijingAir Subseq-50% | MAE: 0.294 -> 0.166 | Improved |
| Add hybrid encoding (ablation) | HELIX · BeijingAir Point-50% | MAE: 0.104 -> 0.102 | Improved |
| Add multi-level fusion (ablation) | HELIX · BeijingAir Block-50% | MAE: 0.147 -> 0.131 | Improved |
| Add multi-level fusion (ablation) | HELIX · ETT-h1 and ItalyAir | MAE | Hurt |
| Add multi-level fusion (ablation) | HELIX · PeMS Subseq-50% | MAE: 0.542 -> 0.311 | Improved |
| Add multi-level fusion (ablation) | HELIX · PeMS Point-10% | MAE: 0.095 -> 0.097 | Hurt |
| Add sinusoidal positional encoding (ablation) | HELIX · BeijingAir Point-50% | MAE: 0.108 -> 0.102 | Improved |
| Replace averaging fusion with learnable gated fusion (ablation) | HELIX · multiple datasets | MAE | Hurt |
| Use HELIX instead of SAITS | imputation baseline · ETT-h1 Point-50% | MAE: .208 -> .189 | Improved |
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.
- CSDI — CSDI: Conditional Score-based Diffusion Models for Probabilistic Time Series Imputation (NeurIPS 2021) · better on 1 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
- Discretizing Continuous Time Series for Imputation with Masked Diffusion Training (NeurIPS 2026 · similarity 0.87)
- ProCTI: Prototype-Refined Global Conditioning for Diffusion-Based Time Series Imputation (NeurIPS 2026 · similarity 0.86)
- Motif-aware Graph Neural Networks for Networked Time Series Imputation (AAAI 2025 · similarity 0.86)
- Graph Structure Learning for Spatial-Temporal Imputation: Adapting to Node and Feature Scales (AAAI 2025 · similarity 0.85)
- CACTI: Leveraging Copy Masking and Contextual Information to Improve Tabular Data Imputation (ICML 2025 · similarity 0.85)
- Propagate and Inject: Revisiting Propagation-Based Feature Imputation for Graphs with Partially Observed Features (ICML 2025 · similarity 0.85)
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.
