TW3Cast: A Frozen Router of Lightly Fine-Tuned Foundation Models for Time-Series Forecasting on GIFT-Eval, Selected Entirely on the Training Split
TW3Cast: A Frozen Router of Lightly Fine-Tuned Foundation Models for Time-Series Forecasting on GIFT-Eval, Selected Entirely on the Training Split
TW3Cast：一种基于训练集完全选定的轻量微调基础模型冻结路由系统，用于 GIFT-Eval 时间序列预测
Abstract: TW3Cast is a time-series forecasting system that reaches position 3 of 130 entries on the GIFT-Eval benchmark by mean MASE rank, as of 2026-09-14. The two entries above it belong to the leaderboard’s agentic category, multi-step systems that use agents or language models to reason about, generate or select forecasts. TW3Cast runs no agent and no language model. Its selection is a table computed once on the training split and then frozen, and its experts are public foundation models lightly fine-tuned on those training splits.
摘要： 截至 2026 年 9 月 14 日，TW3Cast 是一套时间序列预测系统，在 GIFT-Eval 基准测试中以平均 MASE 排名位列 130 个参赛项目中的第 3 名。排在它前面的两个项目属于排行榜的“智能体（agentic）”类别，即使用智能体或语言模型进行推理、生成或选择预测的多步系统。TW3Cast 不运行任何智能体或语言模型。其选择机制是一张在训练集上计算一次后即被冻结的表格，其专家模型是针对这些训练集进行轻量微调的公开基础模型。
For each of the 97 dataset, frequency and horizon configurations, the table serves one of four modes: a specialist, which is a LoRA or full fine-tune of Chronos-2, TiRex or Toto whose training data was cleaned and enriched by explicit rules; a quantile blend that contains a specialist; a blend of base models; or a selection tournament played on a backtest carved from the training split. Every decision in the table was taken on that backtest. A specialist is admitted the moment it beats the tournament there, so a candidate costs a few megabytes and minutes of GPU time, and a failed candidate changes nothing.
针对 97 种数据集、频率和预测范围配置中的每一种，该表格提供四种模式之一：专家模式（对 Chronos-2、TiRex 或 Toto 进行 LoRA 或全量微调，其训练数据通过显式规则进行了清洗和增强）、包含专家的分位数混合模式、基础模型混合模式，或在从训练集划分出的回测中进行的“选择锦标赛”模式。表格中的每一项决策都是基于该回测做出的。一旦某个专家模型在回测中胜出，它就会被采纳，因此每个候选模型的成本仅为几兆字节和几分钟的 GPU 时间，且失败的候选模型不会产生任何影响。
Three guarded mechanisms protect the selection from its own biases: a dual accuracy and calibration criterion, an asymmetric margin against candidates that saw the series during training, and conservative per-window gates. The selection rules themselves were chosen inside a temporal meta-backtest. The best base model served alone reaches a mean MASE rank of 33.8, the tournament served on every configuration reaches 38.0, and the full router reaches 19.4. The routing table, the expert index, the pinned base-model revisions, the submitted score file and the dated snapshot of the public scores are released, and every leaderboard number in this paper regenerates from them by one script.
三种防护机制保护了选择过程免受自身偏差的影响：双重准确率与校准准则、针对在训练期间见过该序列的候选模型的非对称边际，以及保守的窗口级门控。选择规则本身是在一个时间元回测中确定的。表现最好的单一基础模型平均 MASE 排名为 33.8，在所有配置上运行的锦标赛模式排名为 38.0，而完整的路由系统达到了 19.4。路由表、专家索引、固定的基础模型版本、提交的评分文件以及公开评分的日期快照均已发布，本文中的每一个排行榜数据都可以通过一个脚本从这些文件中重新生成。