New EXAONE Demand 1.0 model targets time series forecasting

Researchers have developed EXAONE Demand 1.0, a new time series foundation model specifically designed for demand forecasting. This model utilizes a unique demand-specific corpus of 11.3 million series and 48.4 billion observations, augmented by a synthetic generator to address data properties like short histories and frequent zeros. EXAONE Demand features a demand-aware adapter with low-rank branches for different demand classes and a router to manage their contributions, outperforming 36 existing time series foundation models on held-out datasets. AI

IMPACT This specialized model could improve forecasting accuracy in domains with unique demand characteristics.

RANK_REASON The cluster describes a new research paper detailing a novel AI model for a specific task. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.AI →

• alphaXiv
• arXiv
• CatalyzeX
• DagsHub
• EXAONE Demand
• EXAONE Demand 1.0
• Gotit.pub
• Hugging Face
• ScienceCast
• Time Series Foundation Models

• paper
• model release

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →