Metraipolis › archive › dc_69fece2af6b0
LLMODE: Aligning ODEs with LLMs via Gated Token Injection for Irregular Spatio-Temporal Forecasting — arXiv:2608.29640
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2608.29640
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-17T07:52:11.463Z
- SHA-256 of the archived text
63b3d492068b1ce36edf9e8b16ca78e542b25a01bdb1909ccd1ed9bc96aee871- Stable URL of this page
https://metraipolis.com/documento/dc_69fece2af6b0
How to cite
LLMODE: Aligning ODEs with LLMs via Gated Token Injection for Irregular Spatio-Temporal Forecasting — arXiv:2608.29640 — arXiv (en), https://arxiv.org/abs/2608.29640 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_69fece2af6b0 (Metraipolis archive dc_69fece2af6b0, sha256 63b3d492068b1ce36edf9e8b16ca78e542b25a01bdb1909ccd1ed9bc96aee871)
BibTeX
@misc{dc69fece2af6b0,
title = {LLMODE: Aligning ODEs with LLMs via Gated Token Injection for Irregular Spatio-Temporal Forecasting — arXiv:2608.29640},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2608.29640},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_69fece2af6b0; sha256 63b3d492068b1ce36edf9e8b16ca78e542b25a01bdb1909ccd1ed9bc96aee871},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.29640. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
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. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-01 in cs.AI). Categorie arXiv: Machine Learning (cs.LG) ; Artificial Intelligence (cs.AI). Pagina del lavoro: https://arxiv.org/abs/2608.29640 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-10-17T07:28:26.789Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-01?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).