PaperTwice Upon a Time
Twice Upon a Time: Complementary Representations for Time-Series Reasoning
An academic paper proposing complementary representations for time series reasoning. By combining two visual views, a line plot and a normalized delay embedding, it improves the forecasting and reasoning capabilities of time series language models.
Timeline (last 90 days)
Oct 16
The paper "Twice Upon a Time: Complementary Representations for Time-Series Reasoning" proposes the design principle of providing time-series language models (TSLMs) with two complementary representations simultaneously: line graphs and normalized delay embeddings.
Unverified50%