---
id: 20260607-T0-05
title: "GITCO技术解决时间序列模型预测偏差问题"
title_en: "GITCO Fixes Time Series Model Forecasting Bias"
url: https://ai.daily.yangsir.net/daily/20260607-T0-05
issue_date: 2026-06-07
publish_date: 2026-06-06T04:00:00.000Z
category: research
source_name: "arXiv cs.AI"
source_url: https://arxiv.org/abs/2606.05332
---

# GITCO技术解决时间序列模型预测偏差问题

针对时间序列基础模型TSFMs的预测偏差问题，研究人员提出了GITCO技术。该技术解决了“上下文中毒”现象——异常数据点会过度吸引模型注意力，导致零样本预测质量下降。GITCO通过推理时的门控机制过滤异常上下文，显著提高了预测准确度。实验表明，该方法能有效提升TSFMs在异常数据环境下的鲁棒性，为工业场景中的时间序列预测提供了新方案。

## English Version

**GITCO Fixes Time Series Model Forecasting Bias**

Researchers have developed GITCO to address forecasting bias in Time Series Foundation Models (TSFMs). The technique solves 'context poisoning,' where anomalous patches disproportionately attract model attention, degrading zero-shot forecasts. By using gated mechanisms at inference time to filter anomalous context, GITCO significantly improves prediction accuracy. Experiments show this method enhances TSFMs' robustness in noisy data environments.

---

**来源**：[arXiv cs.AI](https://arxiv.org/abs/2606.05332)

**详情页**：https://ai.daily.yangsir.net/daily/20260607-T0-05

---

*智语观潮 · Daily — https://ai.daily.yangsir.net/llms.txt*