研究人员开发了TSFLab，一个旨在简化时间序列预测（TSF）研究的自改进工作台。该平台解决了已发表TSF论文代码不易获取以及维护发布代码困难等挑战。TSFLab为编码代理提供共享资源和工具，以降低成本并自动化维护，将任务完成率提高高达163%，并显著减少了令牌使用量。此外，还引入了一个名为FABench的新基准环境，通过实现可扩展的回测和即时反馈，超越了依赖日历或固定语料库评估的局限性，从而更有效地评估预测代理。另一个框架FACTS-MAS通过将证据分解为专门的通道并根据预测技能对预测进行加权，提供了一种领域无关的多代理预测方法。 AI
影响 这些在预测框架和基准方面的进展可能会加速跨不同领域的AI驱动的预测能力。
排序理由 多篇研究论文介绍了人工智能驱动预测的新框架和基准。
在 Hugging Face Daily Papers 阅读 →
AI 生成摘要 · Google Gemini · 来自 4 个来源。 我们如何撰写摘要 →
完整方法见我们的编辑标准。
arXiv:2610.40901v1 Announce Type: new Abstract: As publications and products on time series forecasting (TSF) proliferate, researchers face a growing burden of analyzing existing work. First, not all publications release code. For instance, only about 20% of arXiv TSF papers have…
arXiv:2610.40940v1 Announce Type: cross Abstract: Digital agents can now gather and organize information, reason over it, and use tools to complete complex tasks. Forecasting puts all of these abilities to the test: it requires integrating incomplete evidence, inferring causal dr…
arXiv:2610.38225v1 Announce Type: new Abstract: Real-world forecasting often requires integrating heterogeneous evidence whose predictive value varies across horizons, operating regimes, and domains. We introduce FACTS-MAS, a domain-agnostic multi-agent forecasting framework that…
TSFLab is a self-improving agent-agnostic workbench that equips coding agents with shared resources to reduce costs and automate maintenance for time series forecasting research and benchmarking.