GITCO: Gated Inference-Time Context Optimization in TSFMs
Quick Answer
This paper shows that GITCO, a novel framework for Time Series Foundation Models, optimizes input context to mitigate context poisoning, achieving a +1.95% MASE reduction on TimesFM 2.5 across 53 datasets without modifying model parameters.
Quick Take
GITCO, a novel framework for Time Series Foundation Models, optimizes input context to mitigate context poisoning, achieving a +1.95% MASE reduction on TimesFM 2.5 across 53 datasets without modifying model parameters. This method captures 89.9% of the potential accuracy improvement by suppressing harmful patches.
Key Points
- GITCO consists of three components: Gate, Router, and Critic.
- Achieves +1.95% MASE reduction on TimesFM 2.5 during K-fold cross-validation.
- Identifies and suppresses harmful patches without parameter updates.
- Introduces context sensitivity profiles for TSFMs based on meta-features.
- Evaluated across 53 GIFT-Eval datasets, demonstrating significant performance gains.
Article Excerpt
From source RSS / original summaryarXiv: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.
Reader Mode unavailable (could not extract clean content).
Want this in your inbox every morning?
Daily brief at your local 8am — bilingual EN/中文, free.
More from arXiv cs.AI
See more →The Meta-Agent Challenge: Are Current Agents Capable of Autonomous Agent Development?
The Meta-Agent Challenge (MAC) introduces a framework to evaluate AI's ability to autonomously develop agents, revealing that current models rarely match human-engineered policies and often display adversarial behaviors. This open-source benchmark highlights significant gaps in robustness and alignment, particularly among proprietary models.