ICR-Drive: Instruction Counterfactual Robustness for End-to-End Language-Driven Autonomous Driving

• URL: http://arxiv.org/abs/2604.05378v1
• Date: Tue, 07 Apr 2026 03:24:22 GMT
• Title: ICR-Drive: Instruction Counterfactual Robustness for End-to-End Language-Driven Autonomous Driving
• Abstract summary: We introduce ICR-Drive, a diagnostic framework for instruction counterfactual robustness in language-conditioned autonomous driving.<n>We replay identical CARLA routes under matched simulator configurations and seeds to isolate performance changes attributable to instruction language.<n>Experiments on LMDrive and BEVDriver show that minor instruction changes can induce substantial performance drops and distinct failure modes.
• Score: 1.6961250145588112
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Recent progress in vision-language-action (VLA) models has enabled language-conditioned driving agents to execute natural-language navigation commands in closed-loop simulation, yet standard evaluations largely assume instructions are precise and well-formed. In deployment, instructions vary in phrasing and specificity, may omit critical qualifiers, and can occasionally include misleading, authority-framed text, leaving instruction-level robustness under-measured. We introduce ICR-Drive, a diagnostic framework for instruction counterfactual robustness in end-to-end language-conditioned autonomous driving. ICR-Drive generates controlled instruction variants spanning four perturbation families: Paraphrase, Ambiguity, Noise, and Misleading, where Misleading variants conflict with the navigation goal and attempt to override intent. We replay identical CARLA routes under matched simulator configurations and seeds to isolate performance changes attributable to instruction language. Robustness is quantified using standard CARLA Leaderboard metrics and per-family performance degradation relative to the baseline instruction. Experiments on LMDrive and BEVDriver show that minor instruction changes can induce substantial performance drops and distinct failure modes, revealing a reliability gap for deploying embodied foundation models in safety-critical driving.

Related papers

• Latent-Centroid Steering: Single-Pass Classifier-Free Guidance for Command-Aligned Autonomous Driving [37.533498955258544]
  We propose a single-pass guidance mechanism that replaces instance-level residuals with class-level latent shifts.<n>We demonstrate that LCS reduces inference latency by approximately 50% while achieving stronger command adherence.
  arXiv  Detail & Related papers  (2026-07-31T19:32:28Z)
• A Large-Language-Model Supported Personalized Driving Framework for Lane Change in Highway Scenarios [1.6314656571060535]
  This paper proposes a large language model (LLM)-supported personalized driving framework for highway lane-change scenarios.<n>The framework maps natural-language driving commands to executable planning parameters in the open-source Apollo automated driving stack.<n>Experiments show that the derived parameter sets generate distinguishable personalized lane-change behaviors, while RAG consistently improves preference interpretation.
  arXiv  Detail & Related papers  (2026-06-30T10:58:21Z)
• DriveJudge: Rethinking Autonomous Driving Evaluation with Vision-Language Models [53.73898294218455]
  DriveJudge is a driving evaluation agent that combines rule-grounded evaluation with Vision-Language Model (VLM) reasoning.<n>We introduce two benchmark tasks: Driving Quality Classification and Trajectory Preference Selection.<n>DriveJudge outperformsS for driving quality classification by 21.23 AUC, and the recent VLM-based DriveCritic for trajectory preference selection by 6.5%.
  arXiv  Detail & Related papers  (2026-06-15T23:39:36Z)
• OneDrive: Unified Multi-Paradigm Driving with Vision-Language-Action Models [69.2503510410147]
  We present a unified autonomous driving framework built upon a pretrained VLM.<n>We demonstrate that pretrained VLM attention exhibits strong transferability beyond pure language modeling.<n>Experiments on end-to-end autonomous driving benchmarks demonstrate state-of-the-art performance.
  arXiv  Detail & Related papers  (2026-04-20T07:50:00Z)
• Open-Ended Instruction Realization with LLM-Enabled Multi-Planner Scheduling in Autonomous Vehicles [40.195332742819865]
  This study proposes an instruction-realization framework that leverages a large language model (LLM) to interpret instructions.<n>It generates executable scripts that schedule multiple model predictive control (MPC)-based motion planners based on real-time feedback, and converts planned trajectories into control signals.
  arXiv  Detail & Related papers  (2026-04-09T09:32:21Z)
• TAG: Target-Agnostic Guidance for Stable Object-Centric Inference in Vision-Language-Action Models [59.13964209628383]
  Vision-Language--Action (VLA) policies have shown strong progress in mapping language instructions and visual observations to robotic actions, yet their reliability degrades in cluttered scenes with distractors.<n>We propose TAG (Target-Agnostic Guidance), a simple inference-time guidance mechanism that explicitly reduces distractor- and appearance-induced bias in VLA policies.<n>We evaluate TAG on standard manipulation benchmarks, including LIBERO, LIBERO-Plus, and VLABench, where it consistently improves robustness under clutter and reduces near-miss and wrong-object executions.
  arXiv  Detail & Related papers  (2026-03-25T17:56:32Z)
• Unifying Language-Action Understanding and Generation for Autonomous Driving [25.23561391638388]
  Vision-Language-Action (VLA) models are emerging as a promising paradigm for end-to-end autonomous driving.<n>Existing methods suffer from two critical limitations: a persistent misalignment between language instructions and action outputs, and the inherent inefficiency of typical auto-regressive action generation.<n>We introduce LinkVLA, a novel architecture that directly addresses these challenges to enhance both alignment and efficiency.
  arXiv  Detail & Related papers  (2026-03-02T04:41:10Z)
• DRIV-EX: Counterfactual Explanations for Driving LLMs [5.315914253152163]
  Large language models (LLMs) are increasingly used as reasoning engines in autonomous driving.<n>We propose to study their decision process through counterfactual explanations.<n>We introduce DRIV-EX, a method that leverages gradient-based optimization on continuous embeddings.
  arXiv  Detail & Related papers  (2026-02-28T15:12:51Z)
• MindDrive: A Vision-Language-Action Model for Autonomous Driving via Online Reinforcement Learning [51.20229133553804]
  Current Vision-Language-Action (VLA) paradigms in autonomous driving primarily rely on Imitation Learning (IL)<n>Online Reinforcement Learning offers a promising pathway to address these issues through trial-and-error learning.<n>We propose MindDrive, a VLA framework comprising a large language model (LLM) with two distinct sets of LoRA parameters.<n>By feeding trajectory-level rewards back into the reasoning space, MindDrive enables trial-and-error learning over a finite set of discrete linguistic driving decisions.
  arXiv  Detail & Related papers  (2025-12-15T18:31:32Z)
• COMPASS: Context-Modulated PID Attention Steering System for Hallucination Mitigation [2.1521364454860525]
  We introduce a lightweight, interpretable control framework that embeds a model-based feedback loop directly within decoding.<n>We show that a PID controller dynamically modulates attention heads to maintain factual consistency without retraining or multi-pass decoding.
  arXiv  Detail & Related papers  (2025-11-05T05:30:28Z)
• Do What? Teaching Vision-Language-Action Models to Reject the Impossible [53.40183895299108]
  Vision-Language-Action (VLA) models have demonstrated strong performance on a range of robotic tasks.<n>We propose Instruct-Verify-and-Act (IVA), a framework that detects when an instruction cannot be executed due to a false premise.<n>Our experiments show that IVA improves false premise detection accuracy by 97.56% over baselines.
  arXiv  Detail & Related papers  (2025-08-22T10:54:33Z)
• DRIVE: Dynamic Rule Inference and Verified Evaluation for Constraint-Aware Autonomous Driving [37.24058519921229]
  We present DRIVE, a novel framework for Dynamic Rule Inference and Verified Evaluation.<n>D DRIVE achieves 0.0% soft constraint violation rates, smoother trajectories, and stronger generalization across diverse driving scenarios.<n> Verified evaluations further demonstrate the efficiency, explanability, and robustness of the framework for real-world deployment.
  arXiv  Detail & Related papers  (2025-08-06T03:56:06Z)
• DRIVE: Dependable Robust Interpretable Visionary Ensemble Framework in Autonomous Driving [1.4104119587524289]
  Recent advancements in autonomous driving have seen a paradigm shift towards end-to-end learning paradigms. These models often sacrifice interpretability, posing significant challenges to trust, safety, and regulatory compliance. We introduce DRIVE, a comprehensive framework designed to improve the dependability and stability of explanations in end-to-end unsupervised driving models.
  arXiv  Detail & Related papers  (2024-09-16T14:40:47Z)
• Contrastive Instruction Tuning [61.97704869248903]
  We propose Contrastive Instruction Tuning to maximize the similarity between semantically equivalent instruction-instance pairs. Experiments on the PromptBench benchmark show that CoIN consistently improves LLMs' robustness to unseen instructions with variations across character, word, sentence, and semantic levels by an average of +2.5% in accuracy.
  arXiv  Detail & Related papers  (2024-02-17T00:09:32Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.