MetaCaster: Meta-Harness-Optimized Agent for End-to-End Few-Shot Learning of Lightweight Time Series Forecasters
This work proposes MetaCaster, a meta-harness-optimized multi-agent framework that uses agentic data generation to automatically train specialized lightweight forecasters from only a few examples and textual contexts and highlights a new TSF paradigm in which agents act not as forecasters but as intermediary engineers...
Chengao Shen, Wen-Chao Yu, Fang-Yu Wu et al.
· 0 citations