GITCO: Gated Inference-Time Context Optimization in TSFMs

Researchers have developed GITCO, a gated inference-time context optimization framework that improves time series foundation model accuracy by selectively identifying and suppressing harmful data patches without modifying model weights. Tested on TimesFM 2.5 across 53 datasets, GITCO achieved an average 1.95% reduction in forecast error while capturing 89.9% of the theoretical improvement upper bound. The framework also introduces context sensitivity profiles as a new property for characterizing how model architecture and data structure jointly influence potential accuracy gains from inference-time context intervention.

arXiv:2606.05332v1 Announce Type: new Abstract: Patch-based Time Series Foundation Models TSFMs suffer from context poisoning: structurally anomalous patches capture disproportionate attention and silently degrade zero-shot forecast quality. We propose improving TSFM accuracy at inference time by optimizing the input context rather than modifying model weights. We present GITCO Gated Inference-Time Context Optimization , a lightweight three-component framework: Gate, Router, and Critic that selectively identifies and suppresses harmful patches without any parameter updates. Evaluated on TimesFM 2.5 across 53 GIFT-Eval datasets under K-fold cross-validation, GITCO achieves an average +1.95% MASE reduction on TimesFM 2.5 while capturing 89.9% of the improvement upper bound. We introduce context sensitivity profiles as a new characterizable property of TSFMs: the mapping from time series meta-features to expected accuracy improvement under inference-time context intervention, shaped jointly by model architecture and the statistical structure of the data.