Arrow Research search

Author name cluster

YIYANG FU

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

2 papers
1 author row

Possible papers

2

EAAI Journal 2025 Journal Article

A novel dynamic fractional-order discrete grey power model for forecasting China's total solar energy capacity

  • Lin Xia
  • Yuhong Wang
  • Youyang Ren
  • Ke Zhou
  • YIYANG FU

-Precise prediction of the total solar energy capacity is pivotal for the progress of the nation's solar energy industry, the optimization of energy structure and the sustainable development of energy systems. This study proposes a novel dynamic fractional order discrete grey power model (DFDGPM(1, 1)) for predicting China's total solar energy capacity. The model introduces a power exponent to capture the nonlinear characteristics among system behavior variables. Additionally, it incorporates a fractional-order accumulation operator and a dynamic time-delay function, which not only describe the time-delay effect between China's economic development and solar energy growth but also enhance the model's adaptability to different samples. The model demonstrates strong compatibility and can degenerate into 10 existing grey models. Empirical research shows that the model's fitting error is close to 0 %, with a prediction error of only 1. 07 %, which is significantly better than 11 other methods. The forecast findings indicate that China's total solar energy capacity will experience an annual growth rate of 29. 41 % from 2022 to 2030. This method promotes the development of dynamic forecasting technology and provides the necessary technical and data support for renewable energy field.

NeurIPS Conference 2025 Conference Paper

TPP-SD: Accelerating Transformer Point Process Sampling with Speculative Decoding

  • Shukai Gong
  • YIYANG FU
  • Fengyuan Ran
  • Quyu Kong
  • Feng Zhou

We propose TPP-SD, a novel approach that accelerates Transformer temporal point process (TPP) sampling by adapting speculative decoding (SD) techniques from language models. By identifying the structural similarities between thinning algorithms for TPPs and speculative decoding for language models, we develop an efficient sampling framework that leverages a smaller draft model to generate multiple candidate events, which are then verified by the larger target model. TPP-SD maintains the same output distribution as autoregressive sampling while achieving significant acceleration. Experiments on both synthetic and real datasets demonstrate that our approach produces samples from identical distributions as standard methods, but with 2-6$\times$ speedup. Our ablation studies analyze the impact of hyperparameters such as draft length and draft model size on sampling efficiency. TPP-SD bridges the gap between powerful Transformer TPP models and the practical need for rapid sequence generation.

v2026.09.13