Research map / Time series / Long-term forecasting
LMGL-WD: LLM-Guided Multi-Task Graph Learning for Category-Level Warehouse Demand Prediction in E-Commerce
Wenjun Lyu, Fangyu Li, Yudong Zhang, Shuai Wang, Yunhuai Liu, Tian He, Desheng Zhang
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 adaptive similar warehouse selector (cross-warehouse graph) (ablation) | LMGL-WD · Warehouse demand (input 7d, predict next 3d) | MAPE (%): 21.61 -> 19.20 | Improved |
| add cross-category multi-task learning (ablation) | LMGL-WD · Warehouse demand (input 7d, predict next 3d) | MAPE (%): 20.46 -> 19.20 | Improved |
| add adaptive warehouse selector (ablation) | LMGL-WD · Warehouse demand (input 14d, predict next 3d) | MAPE (%): 23.20 -> 20.36 | Improved |
| add cross-category multi-task learning (ablation) | LMGL-WD · Warehouse demand (input 21d, predict next 7d) | MAPE (%): 24.32 -> 22.05 | Improved |
| add cross-category multi-task learning (ablation) | LMGL-WD · Warehouse demand (input 7d, predict next 3d) | MAE: 1.229 -> 1.223 (3d); 1.676 -> 1.682 (7d) | Mixed |
| use LLM-guided category encoding with graph and multi-task modules instead of TimeLLM/TimeCMA | LMGL-WD · Warehouse demand (input 7d, predict next 3d) | MAPE (%): 25.48 -> 19.20 | Improved |
| use LMGL-WD vs best baseline on zero-shot cross-city transfer | LMGL-WD · Cross-city zero-shot (Figure 2) | RMSE, MAE, MAPE: up to 3.08%, 23.18%, 4.37% reductions | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 4 of 4 on most shared results.
- TimeCMA — TimeCMA: Towards LLM-Empowered Multivariate Time Series Forecasting via Cross-Modality Alignment (AAAI 2025) · better on 5 of 5 shared results
- iTransformer — iTransformer: Inverted Transformers Are Effective for Time Series Forecasting (ICLR 2024) · better on 3 of 3 shared results
- TimeLLM — Time-LLM: Time Series Forecasting by Reprogramming Large Language Models (ICLR 2024) · better on 3 of 3 shared results
- AutoFormer — Autoformer: Decomposition Transformers with Auto-Correlation for Long-Term Series Forecasting (NeurIPS 2021) · 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
- EcomGPT: Instruction-Tuning Large Language Models with Chain-of-Task Tasks for E-commerce (AAAI 2024 · similarity 0.82)
- eCeLLM: Generalizing Large Language Models for E-commerce from Large-scale, High-quality Instruction Data (ICML 2024 · similarity 0.82)
- Shopping MMLU: A Massive Multi-Task Online Shopping Benchmark for Large Language Models (NeurIPS 2024 · similarity 0.82)
- A Cross-View Hierarchical Graph Learning Hypernetwork for Skill Demand-Supply Joint Prediction (AAAI 2024 · similarity 0.82)
- GCF: Estimating Unobserved Demand Using Graph Causal Forecasting (AAAI 2025 · similarity 0.80)
- Multi-Domain Deep Learning from a Multi-View Perspective for Cross-Border E-commerce Search (AAAI 2024 · similarity 0.80)
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.
