We lifted 6 functions out of this paper's own repositories and ran 6 of them in a sandbox. "Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results.
| Repository | Role | Ran |
|---|---|---|
| alibaba/aicity2024_track2_aliopentrek_cityllava | canonical | 6 of 6 |
| Function | Status | Where it lives |
|---|---|---|
| enlarge_bbox | Ran | alibaba/aicity2024_track2_aliopentrek_cityllava/data_preprocess/draw_bbox_on_frame.py code served (permissive licence) · get_code("469a3da788c4e23a") |
| enlarge_bbox_square | Ran | alibaba/aicity2024_track2_aliopentrek_cityllava/data_preprocess/draw_bbox_on_frame.py code served (permissive licence) · get_code("e250ce912aa4acd4") |
| find_closest_number | Ran | alibaba/aicity2024_track2_aliopentrek_cityllava/data_preprocess/generate_test_frames.py code served (permissive licence) · get_code("adbd656cbf3f5ab7") |
| get_best_view_wts | Ran | alibaba/aicity2024_track2_aliopentrek_cityllava/data_preprocess/best_view_selection.py code served (permissive licence) · get_code("81ecafec9729489d") |
| load_jsonl | Ran | alibaba/aicity2024_track2_aliopentrek_cityllava/data_preprocess/shortQA_merge.py code served (permissive licence) · get_code("f3a23711c5501382") |
| random_shuffle_conversations | Ran | alibaba/aicity2024_track2_aliopentrek_cityllava/data_preprocess/add_stage_prompt.py code served (permissive licence) · get_code("1f60b7e1b78e5000") |
Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.
In the vast and dynamic landscape of urban settings, Traffic Safety Description and Analysis plays a pivotal role in applications ranging from insurance inspection to accident prevention. This paper introduces CityLLaVA, a novel fine-tuning framework for Visual Language Models (VLMs) designed for urban scenarios. CityLLaVA enhances model comprehension and prediction accuracy through (1) employing bounding boxes for optimal visual data preprocessing, including video best-view selection and visual prompt engineering during both training and testing phases; (2) constructing concise Question-Answer sequences and designing textual prompts to refine instruction comprehension; (3) implementing block expansion to fine-tune large VLMs efficiently; and (4) advancing prediction accuracy via a unique sequential questioning-based prediction augmentation. Demonstrating top-tier performance, our method achieved a benchmark score of 33.4308, securing the leading position on the leaderboard. The code can be found: https://github.com/alibaba/AICITY2024_Track2_AliOpenTrek_CityLLaVA
The same record, over MCP at https://syntology.ai/mcp:
get_harvested_code_for_paper("2405.03194")
get_code_for_paper("2405.03194")
have("2405.03194")
Connect an agent — have() is free.