This week
Suggest a nameObligation Marketing for AI Builders
The 7-Rung LadderThe 8-Type TaxonomyWrapper or WinnerThe Sample ColumnThe Signal
Updated through the day. Every headline links straight to the source. The two lines underneath are ours.
The core finding matters for anyone applying foundation models to specialized time-series problems: zero-shot doesn't work, but fine-tuning does and it's cheap. This is a pattern repeating across vertical tasks. If you're building medical forecasting or domain-specific prediction, spend the week validating your fine-tuning approach instead of betting on foundation model generalization.
This is a contamination study with teeth. Pretrained models win on older test sets but fail on daily exchange rates, suggesting they're learning train-set patterns rather than generalizable forecasting. If you're building financial applications on time-series foundation models, test on real futures, not published benchmarks. The negative result on seasonal strength is worth reading if you're implementing causal filters.
Time series forecasting is a real business problem, and a SOTA model with a commercial license removes friction for enterprise adoption. IBM is positioning Granite as the open-source alternative to proprietary foundation models. If you're building forecasting into a product, this is worth benchmarking against your current stack.
Chaotic systems are hard to predict from short windows, and this paper attempts to give LLMs a phase-space structure that better represents nonlinear dynamics. The idea is reasonable but the execution is complex and the results (in an excerpt) are unshown. Skip this unless you're specifically working on forecasting problems where you have chaotic dynamics and limited data. It's domain-specific and preliminary.