# PhaseShift: Topology-Aware Data Harmonization and Model Consolidation Across Signalized Intersections

arXiv: https://arxiv.org/abs/2608.25275 · HTML: https://mlsanity.com/paper/2608.25275/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 11 of its 22 references that are in the map.
- First posted on arXiv 26 Aug 2026.

## Builds on (most cited first, 11 of 11)

- [VectorNet: Encoding HD Maps and Agent Dynamics from Vectorized Representation](https://mlsanity.com/paper/2005.04259/) (2020)
- [Motion Transformer with Global Intention Localization and Local Movement Refinement](https://mlsanity.com/paper/2209.13508/) (2022)
- [HDGT: Heterogeneous Driving Graph Transformer for Multi-Agent Trajectory Prediction via Scene Encoding](https://mlsanity.com/paper/2205.09753/) (2022)
- [TrafficBots: Towards World Models for Autonomous Driving Simulation and Motion Prediction](https://mlsanity.com/paper/2303.04116/) (2023)
- [Closed-Loop Supervised Fine-Tuning of Tokenized Traffic Models](https://mlsanity.com/paper/2412.05334/) (2024)
- [A Survey on Deep-Learning Approaches for Vehicle Trajectory Prediction in Autonomous Driving](https://mlsanity.com/paper/2110.10436/) (2021)
- [SceneDiffuser++: City-Scale Traffic Simulation via a Generative World Model](https://mlsanity.com/paper/2506.21976/) (2025)
- [Building Transportation Foundation Model via Generative Graph Transformer](https://mlsanity.com/paper/2305.14826/) (2023)
- [IntTrajSim: Trajectory Prediction for Simulating Multi-Vehicle driving at Signalized Intersections](https://mlsanity.com/paper/2506.08957/) (2025)
- [Evaluating Generative Vehicle Trajectory Models for Traffic Intersection Dynamics](https://mlsanity.com/paper/2506.08963/) (2025)
- [Enactor: From Traffic Simulators to Surrogate World Models](https://mlsanity.com/paper/2603.18266/) (2026)

## Abstract (arXiv)

Learned traffic-behavior models are commonly trained separately for each intersection, creating model portfolios that cannot share evidence across sites. We present PhaseShift, a topology-aware framework that harmonizes heterogeneous roadside trajectories into a shared actor-centric representation and trains one reusable backbone. Ego-relative coordinates, trajectory-induced movement paths, normalized signal context, and variable-cardinality interaction tokens remove site conventions while preserving behaviorally relevant topology. The backbone supports pooled operation, zero-shot at a held-out intersection, and low-data adaptation. We evaluate five intersections in two Florida regions on balanced field data, 100k training windows and equal-sized test sets per site under a replay-conditioned, best-of-sampled-trajectory protocol. At 10s, one pooled model lowers both minADE and minFDE relative to trained local models at all five sites, with median reductions of 36.8% and 22.0%. Leave-one-intersection-out deployment, including one cross-region fold, beats local training on both 10-s metrics at four of five sites, although short-horizon performance is less uniform. Fine-tuning with 1,000 target update windows improves on zero-shot at three sites and is the strongest regime at one. At site 7, every cross-site mixture sharply lowers long-horizon error under a fixed 100k-window budget; test-likelihood gains argue against a best-of-sample dispersion-only explanation. Local models fall behind calibrated IDM at the two highest-flow sites after long autoregressive rollouts; pretrained-backbone regimes do not. Within this five-site evaluation, PhaseShift demonstrates consolidation across heterogeneous physical control settings while identifying sites that still require adaptation. The protocol measures conditional single-vehicle generation under replayed context, not closed-loop traffic simulation.