The FalconTST 2.0 model demonstrates State-of-the-Art (SOTA) performance on the Mean Absolute Scaled Error (Metric), one of the most critical metrics used to evaluate time-series models
Ant International has launched Falcon Time-Series Transformer (TST) AI Model 2.0, its newest TST model to deliver more accurate forecasting in real-world FX risk management for cross-border payments. The FalconTST 2.0 model demonstrates State-of-the-Art (SOTA) performance on the Mean Absolute Scaled Error (Metric), one of the most critical metrics used to evaluate time-series models.
The model scored a 0.666 MASE score and surpasses other TST foundational models from world’s other leading technology companies.
Traditional forecasting systems build separate models for different tasks. The FalsonTST model learns common patterns like cycles, trends, sudden shifts, and seasonality from data like finance, energy, and retail. Despite being different industries, there are often commonalities in their underlying temporal structures.
Jiang-Ming Yang, Chief Innovation Officer, Ant International, shared: Large language models have shown how AI can understand and generate information. FalconTST is about another capability that businesses increasingly need: understanding how the world changes over time, and anticipating what comes next.
Kelvin Li, General Manager of Platform Tech and Senior Vice President, Ant International, added: FalconTST helps global businesses — including our own — manage complex cash flow and FX exposure, so they can manage cross-border transactions with greater confidence. With FalconTST 1.0, clients saw real operational value and cost savings from better forecasting.
FalconTST was first deployed internally at Ant International to manage cash flow and FX exposure on an hourly, daily, and weekly basis.
It was later integrated by banks like Barclays, Citi, Deutsche Bank and Standard Chartered into their respective hedging models, with the aim of improving cash flow forecasting and FX liquidity management capabilities.
Comments (0)
Average Rating: No ratings yet/5 (0 reviews)
No comments yet. Be the first to comment!