Papers with time-series forecasting
Revisiting LLMs as Zero-Shot Time Series Forecasters: Small Noise Can Break Large Models (2025.acl-short)
Copied to clipboard
| Challenge: | Large Language Models (LLMs) have shown remarkable performance across diverse tasks without domain-specific training, fueling interest in their potential for time series forecasting. |
| Approach: | They evaluate the effectiveness of LLMs as zero-shot forecasters compared to state-of-the-art domain-specific models by encoding sequences directly within prompts. |
| Outcome: | The proposed models perform well across multiple domains while reducing the need for domain-specific training. |
Time Machine GPT (2024.findings-naacl)
Copied to clipboard
| Challenge: | Large language models are often trained on extensive, temporally indiscriminate text corpora . conventional methods for creating temporal adapted models depend on pre-training static models on time-specific data. |
| Approach: | They propose a series of point-in-time LLMs called TimeMachineGPT to be nonprognosticative . time-series forecasting and event prediction aim to infer a future state from past data . authors propose linguistically-based models that can be used to predict future events . |
| Outcome: | The proposed model is nonprognosticative and ensures it remains uninformed about future factual information and linguistic changes. |
TimeSAF: Towards LLM-Guided Semantic Asynchronous Fusion for Time Series Forecasting (2026.acl-long)
Copied to clipboard
| Challenge: | Existing time series forecasting methods use a deep synchronous fusion strategy . high-level abstract semantics are inappropriately entangled with low-level temporal dynamics . |
| Approach: | They propose a framework based on hierarchical asynchronous fusion that decouples unimodal feature learning from cross-modal interaction. |
| Outcome: | The proposed framework outperforms state-of-the-art approaches on long-term forecasting benchmarks. |