The Case for Negative Data: From Crash Reports to Counterfactuals for Reasonable Driving

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Patrikar, Jay, Sharma, Apoorva, Veer, Sushant, Li, Boyi, Scherer, Sebastian, Pavone, Marco
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909802126376960
author Patrikar, Jay
Sharma, Apoorva
Veer, Sushant
Li, Boyi
Scherer, Sebastian
Pavone, Marco
author_facet Patrikar, Jay
Sharma, Apoorva
Veer, Sushant
Li, Boyi
Scherer, Sebastian
Pavone, Marco
contents Learning-based autonomous driving systems are trained mostly on incident-free data, offering little guidance near safety-performance boundaries. Real crash reports contain precisely the contrastive evidence needed, but they are hard to use: narratives are unstructured, third-person, and poorly grounded to sensor views. We address these challenges by normalizing crash narratives to ego-centric language and converting both logs and crashes into a unified scene-action representation suitable for retrieval. At decision time, our system adjudicates proposed actions by retrieving relevant precedents from this unified index; an agentic counterfactual extension proposes plausible alternatives, retrieves for each, and reasons across outcomes before deciding. On a nuScenes benchmark, precedent retrieval substantially improves calibration, with recall on contextually preferred actions rising from 24% to 53%. The counterfactual variant preserves these gains while sharpening decisions near risk.
format Preprint
id arxiv_https___arxiv_org_abs_2509_18626
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle The Case for Negative Data: From Crash Reports to Counterfactuals for Reasonable Driving
Patrikar, Jay
Sharma, Apoorva
Veer, Sushant
Li, Boyi
Scherer, Sebastian
Pavone, Marco
Robotics
Artificial Intelligence
Learning-based autonomous driving systems are trained mostly on incident-free data, offering little guidance near safety-performance boundaries. Real crash reports contain precisely the contrastive evidence needed, but they are hard to use: narratives are unstructured, third-person, and poorly grounded to sensor views. We address these challenges by normalizing crash narratives to ego-centric language and converting both logs and crashes into a unified scene-action representation suitable for retrieval. At decision time, our system adjudicates proposed actions by retrieving relevant precedents from this unified index; an agentic counterfactual extension proposes plausible alternatives, retrieves for each, and reasons across outcomes before deciding. On a nuScenes benchmark, precedent retrieval substantially improves calibration, with recall on contextually preferred actions rising from 24% to 53%. The counterfactual variant preserves these gains while sharpening decisions near risk.
title The Case for Negative Data: From Crash Reports to Counterfactuals for Reasonable Driving
topic Robotics
Artificial Intelligence
url https://arxiv.org/abs/2509.18626