# Grounding Vision-Language Models in Driving Semantics: A Multi-Dataset Predicate Framework for Explainable Reasoning

arXiv: https://arxiv.org/abs/2609.31636 · HTML: https://mlsanity.com/paper/2609.31636/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 10 of its 17 references that are in the map.
- First posted on arXiv 10 Sep 2026.

## Builds on (most cited first, 10 of 10)

- [nuScenes: A multimodal dataset for autonomous driving](https://mlsanity.com/paper/1903.11027/) (2019)
- [DriveLM: Driving with Graph Visual Question Answering](https://mlsanity.com/paper/2312.14150/) (2023)
- [NuPlan: A closed-loop ML-based planning benchmark for autonomous vehicles](https://mlsanity.com/paper/2106.11810/) (2021)
- [DriveGPT4: Interpretable End-to-end Autonomous Driving via Large Language Model](https://mlsanity.com/paper/2310.01412/) (2023)
- [NuScenes-QA: A Multi-modal Visual Question Answering Benchmark for Autonomous Driving Scenario](https://mlsanity.com/paper/2305.14836/) (2023)
- [LingoQA: Visual Question Answering for Autonomous Driving](https://mlsanity.com/paper/2312.14115/) (2023)
- [Talk2BEV: Language-enhanced Bird's-eye View Maps for Autonomous Driving](https://mlsanity.com/paper/2310.02251/) (2023)
- [NuPlanQA: A Large-Scale Dataset and Benchmark for Multi-View Driving Scene Understanding in Multi-Modal Large Language Models](https://mlsanity.com/paper/2503.12772/) (2025)
- [nuScenes Knowledge Graph -- A comprehensive semantic representation of traffic scenes for trajectory prediction](https://mlsanity.com/paper/2312.09676/) (2023)
- [Relation-based Motion Prediction using Traffic Scene Graphs](https://mlsanity.com/paper/2212.02503/) (2022)

## Abstract (arXiv)

Vision-language models are increasingly used for driving-scene understanding, yet the semantic relations expressed in their outputs are often difficult to verify against the underlying traffic situation. This paper introduces a deterministic multi-dataset predicate framework that derives driving-scene semantics from measurable geometric, kinematic, temporal, map, and traffic-control evidence. Dataset-specific interfaces are used only to recover the required scene information, while predicate definitions remain unchanged across nuPlan and nuScenes and are materialised in a common Predicate Knowledge Graph. Quantitative semantic validation against manually annotated predicate relations on 200 scenarios from each dataset yields macro F1 scores of 0.94 on nuPlan and 0.93 on nuScenes, with an average cross-dataset difference of 0.02 across the shared predicates. The Predicate KG is further evaluated using a frozen LLaVA-OneVision-7B model on the nine NuPlanQA subtasks. Predicate grounding achieves the highest accuracy among the evaluated visual-input conditions in seven of nine NuPlanQA subtasks, including Traffic Light (53.2% to 71.5%), Situation Assessment (76.2% to 86.1%), and Action Recommendation (82.9% to 89.0%). Weather/Lighting remains essentially unchanged (89.4% vs. 88.8%), consistent with the absence of corresponding predicates, while Predicate KG only input outperforms metadata-only input in eight of nine subtasks. The results show that deterministic predicates provide a consistent and traceable semantic representation and, under oracle grounding, can reduce visual dependence for reasoning tasks covered by the predicate vocabulary.