SADWA: Fine-Grained Weather Awareness with Vision-Language Models for Seamless Autonomous Driving in Real Time
Bibliographic record
Abstract
Autonomous driving systems require adaptive driving strategies to ensure safe and efficient operation under diverse weather and road conditions. Traditional weather classification models often categorize images into broad labels such as clear, rainy, or snowy, lacking the finer granularity required for real-world driving scenarios. In this work, we fine-tune a lightweight vision-language model on a dataset that captures nuanced weather conditions, considering factors such as road surface state, precipitation intensity, visibility obstructions, and time of day. Rather than comparing models of different modalities, our focus is to demonstrate that a compact CLIP-based model can efficiently classify 18 strategically defined weather-road interaction classes-providing fast and accurate perception aligned with the demands of autonomous driving. Experimental results show that our approach enables real-time inference (17–19 ms) while maintaining strong classification performance. The proposed method supports practical deployment on resource-constrained platforms and offers a scalable path toward fine-grained weather awareness for autonomous systems, ultimately enhancing safety and decision-making under adverse environmental conditions.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.002 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.002 |
| Open science | 0.002 | 0.001 |
| Research integrity | 0.001 | 0.002 |
| Insufficient payload (model declined to judge) | 0.002 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".