---
title: "Structured Representation Probing for Risk Anticipation in End-to-End Autonomous Driving"
source: "arXiv AI/CL/LG（论文）"
source_tier: "t2"
url: "https://arxiv.org/abs/2610.37299v1"
permalink: "https://feed.astralor.com/items/itm_108ec8cb52b1711340b90340"
published: "2026-10-27T14:20:15.000Z"
score: 53
channel: "ai"
tags: ["多模态","Structured Representation Probing","Risk Anticipation","End-to-End Autonomous Driving","End-to-end","These","Structured Representation Probe"]
---

# Structured Representation Probing for Risk Anticipation in End-to-End Autonomous Driving

## 速览

End-to-end driving policies form structured internal representations of traffic, ego behavior, and road constraints. These representations offer a monitoring interface for anticipating failures, complementing risk signa…

## 关键点

- End-to-end driving policies form structured internal representations of traffic, ego behavior, and road constraints.
- These representations offer a monitoring interface for anticipating failures, complementing risk signals from trajectories, controls, and manually designed features.
- We introduce the Structured Representation Probe (SRP), a lightweight monitor that reads actor, ego, and map representations from a frozen policy.

## 全文

End-to-end driving policies form structured internal representations of traffic, ego behavior, and road constraints. These representations offer a monitoring interface for anticipating failures, complementing risk signals from trajectories, controls, and manually designed features. We introduce the Structured Representation Probe (SRP), a lightweight monitor that reads actor, ego, and map representations from a frozen policy. SRP learns a separate linear risk direction for each source and aggregates element scores through cardinality-normalized log-mean-exp pooling, retaining sensitivity to salient evidence while reducing dependence on token count. Event-window supervision links the readout to upcoming hard failures, and auxiliary event type heads jointly train the source directions across failure modes. Online alerting uses a single generic score without modifying the driving policy. Bench2Drive experiments on DriveTransformer, UniAD, and VAD show higher frame-level PR-AUC and ROC-AUC than the evaluated runtime baselines. On DriveTransformer at a horizon of 2 seconds, ROC-AUC reaches 0.833 versus 0.715 for temporal consistency, with 2.67% readout runtime overhead. The trained probes also retain the ability to distinguish impending failures on new routes.

---

引用前请核对原文;本文由星际流动整理:https://feed.astralor.com/items/itm_108ec8cb52b1711340b90340