5 papers
Parallel Predictive World Models for Accurate and Efficient Long-Horizon Planning
Wanjin Feng, Baobin Zhang, Ao Yu +3
Long-horizon world-model planning typically relies on autoregressive rollouts, where predicted states are repeatedly fed back into the model. This preserves temporal structure but…
Beyond Model Ranking: Predictability-Aligned Evaluation for Time Series Forecasting
Wanjin Feng, Yuan Yuan, Jingtao Ding +1
In the era of increasingly complex AI models for time series forecasting, progress is often measured by marginal improvements on benchmark leaderboards. However, this approach suff…
Efficient Parallel Training Methods for Spiking Neural Networks with Constant Time Complexity
Wanjin Feng, Xingyu Gao, Wenqian Du +4
Spiking Neural Networks (SNNs) often suffer from high time complexity due to the sequential processing of spikes, making training computationally expensive. In this pape…
TS-LIF: A Temporal Segment Spiking Neuron Network for Time Series Forecasting
Shibo Feng, Wanjin Feng, Xingyu Gao +2
Spiking Neural Networks (SNNs) offer a promising, biologically inspired approach for processing spatiotemporal data, particularly for time series forecasting. However, conventional…
GiFT: Gibbs Fine-Tuning for Code Generation
Haochen Li, Wanjin Feng, Xin Zhou +1
Training Large Language Models (LLMs) with synthetic data is a prevalent practice in code generation. A key approach is self-training, where LLMs are iteratively trained on self-ge…