Recently the performance of long-term time series forecasting has been greatly improved by deep models. In this paper, we propose a general framework called Multi-scale Moving Transformation (MMT) that can be applied to the state-of-the-art models of time series forecasting tasks. Via comparative experiments, we demonstrate the effectiveness of the MMT framework against the models with the moving average. By incorporating the MMT framework with state-of-the-art backbone models based on different methods, our experimental results on various public datasets demonstrate that the improvements outperform their corresponding baseline counterparts.

Conversational text-to-SQL is designed to translate multi-turn natural language questions into their corresponding SQL queries. Most advanced conversational text-to-SQL methods are incompatible with generative pre-trained language models (PLMs), such as T5. In this paper, we present a two-stage unified MultI-task Generation frAmework (MIGA) that leverages PLMs’ ability to tackle conversational text-to-SQL. In the pre-training stage, MIGA first decomposes the main task into several related sub-tasks and then unifies them into the same sequence-to-sequence (Seq2Seq) paradigm with task-specific natural language prompts to boost the main task from multi-task training. Later in the fine-tuning stage, we propose four SQL perturbations to alleviate the error propagation problem. MIGA tends to achieve state-of-the-art performance on two benchmarks (SparC and CoSQL). We also provide extensive analyses and discussions to shed light on some new perspectives for conversational text-to-SQL.