# GContextFormer: A global context-aware hybrid multi-head attention approach with scaled additive aggregation for multimodal trajectory prediction

arXiv: https://arxiv.org/abs/2511.18874 · HTML: https://mlsanity.com/paper/2511.18874/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 19 of its 43 references that are in the map.
- First posted on arXiv 24 Nov 2025.

## Builds on (most cited first, 19 of 19)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [Neural Machine Translation by Jointly Learning to Align and Translate](https://mlsanity.com/paper/1409.0473/) (2014)
- [Social-STGCNN: A Social Spatio-Temporal Graph Convolutional Neural Network for Human Trajectory Prediction](https://mlsanity.com/paper/2002.11927/) (2020)
- [VectorMapNet: End-to-end Vectorized HD Map Learning](https://mlsanity.com/paper/2206.08920/) (2022)
- [MapTRv2: An End-to-End Framework for Online Vectorized HD Map Construction](https://mlsanity.com/paper/2308.05736/) (2023)
- [GroupNet: Multiscale Hypergraph Neural Networks for Trajectory Prediction with Relational Reasoning](https://mlsanity.com/paper/2204.08770/) (2022)
- [Multimodal Trajectory Prediction Conditioned on Lane-Graph Traversals](https://mlsanity.com/paper/2106.15004/) (2021)
- [Social-Implicit: Rethinking Trajectory Prediction Evaluation and The Effectiveness of Implicit Maximum Likelihood Estimation](https://mlsanity.com/paper/2203.03057/) (2022)
- [Producing and Leveraging Online Map Uncertainty in Trajectory Prediction](https://mlsanity.com/paper/2403.16439/) (2024)
- [Enhancing Vectorized Map Perception with Historical Rasterized Maps](https://mlsanity.com/paper/2409.00620/) (2024)
- [GATraj: A Graph- and Attention-based Multi-Agent Trajectory Prediction Model](https://mlsanity.com/paper/2209.07857/) (2022)
- [Trajectory Prediction for Autonomous Driving: Progress, Limitations, and Future Directions](https://mlsanity.com/paper/2503.03262/) (2025)
- [Context-Aware Timewise VAEs for Real-Time Vehicle Trajectory Prediction](https://mlsanity.com/paper/2302.10873/) (2023)
- [MFTraj: Map-Free, Behavior-Driven Trajectory Prediction for Autonomous Driving](https://mlsanity.com/paper/2405.01266/) (2024)
- [Towards Explainable Motion Prediction using Heterogeneous Graph Representations](https://mlsanity.com/paper/2212.03806/) (2022)
- [Multiclass-SGCN: Sparse Graph-based Trajectory Prediction with Agent Class Embedding](https://mlsanity.com/paper/2206.15275/) (2022)
- [Vision-based Multi-future Trajectory Prediction: A Survey](https://mlsanity.com/paper/2302.10463/) (2023)
- [SocialFormer: Social Interaction Modeling with Edge-enhanced Heterogeneous Graph Transformers for Trajectory Prediction](https://mlsanity.com/paper/2405.03809/) (2024)
- [Map-Free Trajectory Prediction with Map Distillation and Hierarchical Encoding](https://mlsanity.com/paper/2411.10961/) (2024)

## Abstract (arXiv)

Multimodal trajectory prediction generates multiple plausible future trajectories to address vehicle motion uncertainty from intention ambiguity and execution variability. However, HD map-dependent models suffer from costly data acquisition, delayed updates, and vulnerability to corrupted inputs, causing prediction failures. Map-free approaches lack global context, with pairwise attention over-amplifying straight patterns while suppressing transitional patterns, resulting in motion-intention misalignment. This paper proposes GContextFormer, a plug-and-play encoder-decoder architecture with global context-aware hybrid attention and scaled additive aggregation achieving intention-aligned multimodal prediction without map reliance. The Motion-Aware Encoder builds scene-level intention prior via bounded scaled additive aggregation over mode-embedded trajectory tokens and refines per-mode representations under shared global context, mitigating inter-mode suppression and promoting intention alignment. The Hierarchical Interaction Decoder decomposes social reasoning into dual-pathway cross-attention: a standard pathway ensures uniform geometric coverage over agent-mode pairs while a neighbor-context-enhanced pathway emphasizes salient interactions, with gating module mediating their contributions to maintain coverage-focus balance. Experiments on eight highway-ramp scenarios from TOD-VT dataset show GContextFormer outperforms state-of-the-art baselines. Compared to existing transformer models, GContextFormer achieves greater robustness and concentrated improvements in high-curvature and transition zones via spatial distributions. Interpretability is achieved through motion mode distinctions and neighbor context modulation exposing reasoning attribution. The modular architecture supports extensibility toward cross-domain multimodal reasoning tasks. Source: https://fenghy-chen.github.io/sources/.