Time-R1 trains LLMs via supervised fine-tuning followed by reinforcement learning with a time-series-specific reward and non-uniform GRIP sampling to enable multi-step reasoning that improves forecasting accuracy.
Using July 4th as a base, each subsequent day’s hourly values are incremented by small amounts (0.1–0.3 or 2–5%), more for peak hours
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Time Series Forecasting as Reasoning: A Slow-Thinking Approach with Reinforced LLMs
Time-R1 trains LLMs via supervised fine-tuning followed by reinforcement learning with a time-series-specific reward and non-uniform GRIP sampling to enable multi-step reasoning that improves forecasting accuracy.