DiffVLA++: Bridging Cognitive Reasoning and End-to-End Driving through Metric-Guided Alignment

TL;DR

DiffVLA++ is a novel autonomous driving framework that combines cognitive reasoning and end-to-end planning through metric-guided alignment, improving generalization and physical feasibility in complex scenarios.

Contribution

It introduces a VLA module for semantic trajectories, an E2E module with a dense trajectory vocabulary, and a metric-guided scorer to align their outputs, enhancing autonomous driving performance.

Findings

Achieves an EPDMS of 49.12 on the ICCV 2025 Autonomous Grand Challenge leaderboard.

Effectively combines world knowledge and physical reasoning for robust driving.

Demonstrates improved handling of long-tail and challenging scenarios.

Abstract

Conventional end-to-end (E2E) driving models are effective at generating physically plausible trajectories, but often fail to generalize to long-tail scenarios due to the lack of essential world knowledge to understand and reason about surrounding environments. In contrast, Vision-Language-Action (VLA) models leverage world knowledge to handle challenging cases, but their limited 3D reasoning capability can lead to physically infeasible actions. In this work we introduce DiffVLA++, an enhanced autonomous driving framework that explicitly bridges cognitive reasoning and E2E planning through metric-guided alignment. First, we build a VLA module directly generating semantically grounded driving trajectories. Second, we design an E2E module with a dense trajectory vocabulary that ensures physical feasibility. Third, and most critically, we introduce a metric-guided trajectory scorer that guides and aligns the outputs of the VLA and E2E modules, thereby integrating their complementary strengths. The experiment on the ICCV 2025 Autonomous Grand Challenge leaderboard shows that DiffVLA++ achieves EPDMS of 49.12.