• Time series forecasting has undergone a transformation with the emergence of foundation models, moving beyond traditional statistical methods that extrapolate from single time series. • Building on the success of the original Chronos models - which have been downloaded over 600 million times from Hugging Face - Amazon researchers introduce Chronos-2, designed to handle arbitrary forecasting tasks in a zero-shot manner through in-context learning (ICL). • Unlike its predecessors, which supported only univariate forecasting, Chronos-2 can jointly predict multiple coevolving time series (multivariate forecasting) and incorporate external factors like promotional schedules or weather conditions (covariate-informed forecasting). • For example, cloud operations teams can forecast CPU usage, memory consumption, and storage I/O together, while retailers can factor in planned promotions when predicting demand. • The model’s group attention mechanism enables it to capture complex interactions between variables, making it particularly valuable for cold-start scenarios where limited historical data is available. • Quantum computing has long promised exponentially faster computation for certain problems, but quantum devices’ extreme sensitivity to environmental noise has limited practical applications.

Article Summaries:

  • Amazon announced several high‑profile advances in 2025. The new Chronos‑2 foundation model extends its predecessor’s zero‑shot forecasting to multivariate and covariate‑informed tasks, using group attention to capture interactions among variables and enabling cold‑start predictions. In quantum computing, AWS’s Ocelot chip introduces bosonic “cat‑qubit” error correction, achieving bit‑flip times near one second and a distance‑5 code with only nine qubits-far fewer than the 49 required by conventional surface codes. Researchers also explored agentic AI, proposing neural‑embedding communication between agents and balanced context sharing for privacy. Finally, Amazon released Mitra, a tabular foundation model trained exclusively on synthetic data.

Sources: