Saved in:
Bibliographic Details
Main Authors: Goldshmidt, Roni, Scott, Hamish, Niccolini, Lorenzo, Matzner, Hernan
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2604.05767
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866918440682389504
author Goldshmidt, Roni
Scott, Hamish
Niccolini, Lorenzo
Matzner, Hernan
author_facet Goldshmidt, Roni
Scott, Hamish
Niccolini, Lorenzo
Matzner, Hernan
contents We present BADAS-2.0, the second generation of our collision anticipation system, building on BADAS-1.0, which showed that fine-tuning V-JEPA2 on large-scale ego-centric dashcam data outperforms both academic baselines and production ADAS systems. BADAS-2.0 advances the state of the art along three axes. (i) Long-tail benchmark and accuracy: We introduce a 10-group long-tail benchmark targeting rare and safety-critical scenarios. To construct it, BADAS-1.0 is used as an active oracle to score millions of unlabeled drives and surface high-risk candidates for annotation. Combined with Nexar's Atlas platform for targeted data collection, this expands the dataset from 40k to 178,500 labeled videos (~2M clips), yielding consistent gains across all subgroups, with the largest improvements on the hardest long-tail cases. (ii) Knowledge distillation to edge: Domain-specific self-supervised pre-training on 2.25M unlabeled driving videos enables distillation into compact models, BADAS-2.0-Flash (86M) and BADAS-2.0-Flash-Lite (22M), achieving 7-12x speedup with near-parity accuracy, enabling real-time edge deployment. (iii) Explainability: BADAS-2.0 produces real-time object-centric attention heatmaps that localize the evidence behind predictions. BADAS-Reason extends this with a vision-language model that consumes the last frame and heatmap to generate driver actions and structured textual reasoning. Inference code and evaluation benchmarks are publicly available.
format Preprint
id arxiv_https___arxiv_org_abs_2604_05767
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Beyond the Beep: Scalable Collision Anticipation and Real-Time Explainability with BADAS-2.0
Goldshmidt, Roni
Scott, Hamish
Niccolini, Lorenzo
Matzner, Hernan
Computer Vision and Pattern Recognition
Computation and Language
We present BADAS-2.0, the second generation of our collision anticipation system, building on BADAS-1.0, which showed that fine-tuning V-JEPA2 on large-scale ego-centric dashcam data outperforms both academic baselines and production ADAS systems. BADAS-2.0 advances the state of the art along three axes. (i) Long-tail benchmark and accuracy: We introduce a 10-group long-tail benchmark targeting rare and safety-critical scenarios. To construct it, BADAS-1.0 is used as an active oracle to score millions of unlabeled drives and surface high-risk candidates for annotation. Combined with Nexar's Atlas platform for targeted data collection, this expands the dataset from 40k to 178,500 labeled videos (~2M clips), yielding consistent gains across all subgroups, with the largest improvements on the hardest long-tail cases. (ii) Knowledge distillation to edge: Domain-specific self-supervised pre-training on 2.25M unlabeled driving videos enables distillation into compact models, BADAS-2.0-Flash (86M) and BADAS-2.0-Flash-Lite (22M), achieving 7-12x speedup with near-parity accuracy, enabling real-time edge deployment. (iii) Explainability: BADAS-2.0 produces real-time object-centric attention heatmaps that localize the evidence behind predictions. BADAS-Reason extends this with a vision-language model that consumes the last frame and heatmap to generate driver actions and structured textual reasoning. Inference code and evaluation benchmarks are publicly available.
title Beyond the Beep: Scalable Collision Anticipation and Real-Time Explainability with BADAS-2.0
topic Computer Vision and Pattern Recognition
Computation and Language
url https://arxiv.org/abs/2604.05767