Problem
Research questions and friction points this paper is trying to address.
spatio-temporal forecasting
irregular observations
temporal asynchrony
representation-space misalignment
limited context windows
LLMODE: Aligning ODEs with LLMs via Gated Token Injection for Irregular Spatio-Temporal Forecasting
🤖 AI Summary
LLMODE通过图感知ODE编码器和固定预算感知重采样器解决不规则时空预测问题，将数据转换为连续时间潜在轨迹并注入冻结的大型语言模型。
📝 Abstract
Large language models (LLMs) have shown promise for spatio-temporal forecasting, but existing approaches often rely on regularly sampled token sequences and struggle with irregular observations because of temporal asynchrony, representation-space misalignment, and limited context windows. We propose LLMODE, a token-efficient framework for irregular spatio-temporal forecasting with a frozen LLM backbone. LLMODE first uses a graph-aware ODE encoder to reconstruct irregular graph observations as a continuous-time latent trajectory. A Fixed-Budget Perceiver Resampler then compresses this variable-length trajectory into a fixed number of dynamic memory tokens. In parallel, compact statistical descriptors are encoded and resampled into context memory tokens. A dual-source gated cross-attention module injects both memories into the frozen LLM, enabling controlled utilization of external spatio-temporal evidence. Experiments on three real-world urban datasets and two physical-dynamics benchmarks show competitive overall performance, with clearer advantages under sparse or dynamically complex irregular sampling. Additional evaluations on unseen urban regions further demonstrate strong zero-shot generalization without adaptation.
Innovation
Methods, ideas, or system contributions that make the work stand out.
Gated Token Injection
Graph-aware ODE Encoder
Fixed-Budget Perceiver Resampler
Dual-source Gated Cross-attention
🔎 Similar Papers
