Research map / Time series / Foundation models for time series

Breakthrough Sensor-Limited Single View: Towards Implicit Temporal Dynamics for Time Series Domain Adaptation

NeurIPS 2025 · Time series › Foundation models for time series · approach: Time series representation learning · 0 citations (OpenAlex)

Read the paperCheck your idea against 108,967 papers

AI-extracted summary. Written by AI from the paper's abstract only. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Proposes an approach to unsupervised time series domain adaptation that models implicit latent continuous-time dynamics that sensor sampling only partially reveals through discrete time series.

Problem

Unsupervised domain adaptation for time series is hard because domain shifts are entangled with intricate temporal patterns, and the latent continuous-time dynamics are only partially observable through discrete, sensor-limited samples.

Approach

Treats the latent continuous-time dynamics as implicit and recovers them from discrete observations to align source and target domains, rather than relying only on the observed discrete sequence. Full method details are unavailable beyond the abstract.

Closest papers by meaning

• CauDiTS: Causal Disentangled Domain Adaptation of Multivariate Time Series (ICML 2024 · similarity 0.89)
• Boosting Transferability and Discriminability for Time Series Domain Adaptation (NeurIPS 2024 · similarity 0.89)
• Enhancing Multivariate Time-Series Domain Adaptation via Contrastive Frequency Graph Discovery and Language-Guided Adversary Alignment (AAAI 2025 · similarity 0.88)
• Large Pre-trained time series models for cross-domain Time series analysis tasks (NeurIPS 2024 · similarity 0.88)
• Efficient Source-Free Time-Series Adaptation via Parameter Subspace Disentanglement (ICLR 2025 · similarity 0.87)
• Learning Pattern-Specific Experts for Time Series Forecasting Under Patch-level Distribution Shift (NeurIPS 2025 · similarity 0.87)