# AutoMine Solution for AV2 2026 Scenario Mining Challenge

arXiv: https://arxiv.org/abs/2606.11874 · HTML: https://mlsanity.com/paper/2606.11874/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 5 of its 6 references that are in the map.
- First posted on arXiv 10 Jun 2026.

## Builds on (most cited first, 5 of 5)

- [Quantifying Language Models' Sensitivity to Spurious Features in Prompt Design or: How I learned to start worrying about prompt formatting](https://mlsanity.com/paper/2310.11324/) (2023)
- [Argoverse 2: Next Generation Datasets for Self-Driving Perception and Forecasting](https://mlsanity.com/paper/2301.00493/) (2023)
- [Immortal Tracker: Tracklet Never Dies](https://mlsanity.com/paper/2111.13672/) (2021)
- [RefAV: Towards Planning-Centric Scenario Mining](https://mlsanity.com/paper/2505.20981/) (2025)
- [Technical Report for Argoverse Challenges on Unified Sensor-based Detection, Tracking, and Forecasting](https://mlsanity.com/paper/2311.15615/) (2023)

## Abstract (arXiv)

With the development of autonomous driving systems, mining high-value, safety-critical, and planning-relevant scenarios from large-scale driving logs has become essential for data-driven evaluation. In this paper, we propose AutoMine, a robust self-refining scenario mining method based on LLMs and VLMs. AutoMine uses semantics-preserving prompt augmentation to reduce LLM prompt sensitivity, combines robust trajectory atomic functions with VLM-based functions to handle perception noise and open-world visual cues, and refines generated code through execution feedback from real logs. In the Argoverse 2 Scenario Mining Competition at CVPR 2026, AutoMine achieves a HOTA-Temporal score of 36.38 and a Timestamp BA score of 77.21.