Lintao Ma

Researcher@Ant Group·中国·OpenReview

研究方向

Probabilistic Methods · Latent Variable Models · Variational Inference · Time Series Analysis · Quantitative Finance and Econometrics · Deep Learning · Generative Models · Memory-Augmented Neural Networks · Attention Models

5.5

UniCA: Unified Covariate Adaptation for Time Series Foundation Model

ICLR 2026Poster

3.5

Kairos: Towards Adaptive and Generalizable Time Series Foundation Models

ICLR 2026Rejected

3.5

Attention Localization Through Separator Tokens: Unlocking Long Numerical Sequence Processing in LLMs

ICLR 2026Withdrawn

3.3

Xihe: Scalable Zero-shot Time Series Learner via Hierarchical Interleaved Block Attention

ICLR 2026Rejected

通讯

3.2

Useless, or Untapped? Unlocking the Full Value of Zero-Advantage Samples for Better Policy Optimization

ICLR 2026Withdrawn

6.8

TS-RAG: Retrieval-Augmented Generation based Time Series Foundation Models are Stronger Zero-Shot Forecaster

NeurIPS 2025Poster

5.5

TS-RAG: Retrieval-Augmented Generation based Time Series Foundation Models are Stronger Zero-Shot Forecaster

合作者 (20)