Introduces structured NuScenes-S dataset and 0.9B FastDrive VLM claiming 20% higher decision accuracy and over 10x inference speedup versus larger unstructured VLMs.
Qwen2.5 technical report,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2025 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Structured Labeling Enables Faster Vision-Language Models for End-to-End Autonomous Driving
Introduces structured NuScenes-S dataset and 0.9B FastDrive VLM claiming 20% higher decision accuracy and over 10x inference speedup versus larger unstructured VLMs.