Notice: The reproducibility variables underlying each score are classified using an automated LLM-based pipeline, validated against a manually labeled dataset. LLM-based classification introduces uncertainty and potential bias; scores should be interpreted as estimates. Full accuracy metrics and methodology are described in Coakley et alK. L. Coakley, T. Snelleman, H. Hoos, and O. E. Gundersen, "The embrace of open science: An analysis of a decade of AI research and 56 800 conference papers," Under Review, 2026..

A Driving-Style-Adaptive Framework for Vehicle Trajectory Prediction

Authors: Di Wen, Yu Wang, Zhigang Wu, Zhaocheng He, Zhe Wu, Zheng Qingfang

NeurIPS 2025 | Venue PDF | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Extensive experiments on public real-world datasets demonstrate that the DSA framework outperforms state-of-the-art methods.
Researcher Affiliation Academia 1Sun Yat-sen University 2Pengcheng Laboratory 3Guangdong Provincial Key Laboratory of Intelligent Transportation System EMAIL EMAIL EMAIL
Pseudocode Yes Algorithm: Polynomial Combination
Open Source Code Yes We provide the code in supplemental material.
Open Datasets Yes Extensive experiments on public real-world datasets demonstrate that the DSA framework outperforms state-of-the-art methods. ... We evaluate our DSA framework on three real-world vehicle trajectory prediction datasets: nu Scenes [67], Argoverse [68] and Waymo [69].
Dataset Splits No Specifically, the degree n is treated as a hyperparameter optimization problem, aimed at minimizing the loss (L) on validation data Dval and training data Dtrain. While Dtrain and Dval are mentioned, the specific split percentages or methodology for these splits are not provided in the paper.
Hardware Specification No No specific hardware details (like GPU/CPU models, memory) are provided in the paper.
Software Dependencies No To tackle this issue, we utilize SMAC3 [66] tool, which is particularly suitable for optimizing low-dimensional and continuous functions... However, no specific version number for SMAC3 or other main software dependencies are provided.
Experiment Setup Yes We utilize Loss = λ1LDis + λ2LMo E-K, with LMo E-K = wload CV (loads)2, for model training with balanced weighting parameters λ . ... Specifically, the degree n is treated as a hyperparameter optimization problem, aimed at minimizing the loss (L) on validation data Dval and training data Dtrain.