- Rating
- 2
- Confidence
- 3
- Informativeness
- 2
- Presentation
- 3
- Soundness
- 2
Read review
Paper summary
This paper studies the optimization challenges of token-native VLAs for autonomous driving. The authors argue that cross-entropy loss does not account for the physical distance between discrete action tokens. As a result, predictions with different trajectory errors may receive the same loss, leading to poor closed-loop driving performance.
To address this issue, the authors adapt Ordered Action Tokenization (OAT) to driving trajectories, using coarse-to-fine token sequences to establish a relationship between shared prefixes and spatial distance. They further propose Coarse Trajectory Anchor Regularization (CTAR), which introduces continuous spatial distance supervision for the first coarse token through a frozen trajectory decoder.
Experiments on CARLA Bench2Drive and NAVSIM v2 show substantial improvements over K-DISK. The proposed method also matches or outperforms the evaluated continuous control baselines on the overall metrics.
Critical strengths
-
The paper addresses a meaningful problem. Rather than simply modifying the model architecture, the authors focus on the mismatch between classification loss and physical distance in discrete action spaces. In particular, K-DISK achieves lower trajectory reconstruction error but higher actual prediction error, which provides strong motivation for the proposed approach.
-
The method is well motivated and shows strong experimental results. OAT and CTAR address discrete action prediction from two complementary perspectives: ordered action representations and continuous distance supervision. The method achieves clear improvements on two autonomous driving benchmarks, and the CTAR ablation demonstrates the importance of this component.
Critical weaknesses
-
Inference efficiency is not quantitatively evaluated. The proposed method requires autoregressive generation of 32 action tokens, which may introduce substantial inference overhead compared with a continuous regression head. Although the paper discusses the impact of inference latency on evaluation, it does not report actual inference time or throughput. Given the real-time requirements of autonomous driving, a latency comparison with the continuous baseline is needed to assess the trade-off between performance gains and inference cost.
-
Some experimental results are not sufficiently analyzed. On NAVSIM v2, although the proposed method achieves a higher EPDMS, its Extended Comfort score drops substantially from 0.747 for the continuous baseline to 0.373. The paper does not adequately explain this degradation.
Questions for authors
-
What are the actual inference latency and throughput when generating 32 action tokens autoregressively? Compared with the continuous regression baseline, can the proposed method meet the same real-time control requirements?
-
Why does OAT + CTAR achieve a substantially lower Extended Comfort score than the continuous baseline? Is this degradation related to trajectory smoothness, acceleration, or jerk? It would be helpful to provide further analysis of the underlying causes.
-
CTAR applies distance supervision only to the first coarse token, while cross-entropy for the subsequent tokens still lacks explicit spatial distance information. Could the authors further explain the rationale for supervising only the first token? Could applying distance supervision to multiple prefixes further improve trajectory quality?
3 comments
Please help to join the discussion
Dear Reviewer nBkZ,
The discussion deadline is approaching. Please help to join the discussion and offer the reviewer's opportunity to further respond to your concerns.
Thanks!
AC
Thank you for the detailed response. The analysis of cross-frame inconsistency and longitudinal jerk provides a reasonable explanation for the drop in Extended Comfort. The rationale for applying CTAR only to the first token is also convincing. My main concern remains inference efficiency. Although the authors provide additional results on prefix truncation, actual latency and throughput measurements are still missing, making it difficult to assess the overhead relative to continuous regression heads and the trade-off between performance and efficiency. Overall, I appreciate the value of the problem and the experimental results. However, as inference efficiency remains insufficiently evaluated, I will maintain my original rating.
We thank the reviewer for the thoughtful questions and constructive feedback. We address each point below and have updated the revised PDF accordingly:
1. Inference Latency, Throughput, and Real-Time Control (Weakness 1 & Question 1): We acknowledge that our study does not benchmark wall-clock latency or throughput: neither OAT + CTAR nor the compared VLM baselines are real-time systems in their current configurations, as all share the same
InternVL2-1Bbackbone processing 1,024 visual tokens plus autoregressive language tokens before action emission.InternVL2-1Bvisual prefill and text generation.scenario44(35.91vs.29.65 DS; Section 4.5).2. Root Cause of Lower Extended Comfort (
0.373vs.0.747) on NAVSIM v2 (Weakness 2 & Question 2): We explicitly analyze this degradation as a limitation in Sections 4.3 and 5 (alongside Appendix Section A.9):C = 0.962vs.0.976; Table 2), with zero lateral acceleration inflation (1.0xbaseline) and superior spatial compliance (DAC = 0.853vs.0.753,LC = 0.964vs.0.940,TTC = 0.942vs.0.931). Individual trajectories decoded at each tick are physically smooth.C,ECmeasures temporal consistency across consecutive replanning frames. Because our policy decodes memorylessly at each tick without conditioning on the prior frame's predicted prefix, minor visual shifts near a codebook boundary can step argmax selection between adjacent discrete codes. As shown in Appendix Section A.9, our deployed NAVSIM codec (1.182 mADE) has residual quantization error that is along-track dominated (2.1–3.1xalong-to-cross track ratio). Discrete longitudinal position/velocity jumps between frames t and t+1 induce6.4–13.6xlongitudinal jerk inflation across ticks, penalizingEC; whereas continuous heads vary smoothly with small visual perturbations.73.75 DS,56.4%SR; Table 1). Prior-tick action prefix conditioning and cross-frame smoothness losses are highlighted as future work.3. Rationale for Supervising Only t=1 in CTAR & Multi-Prefix Extension (Question 3):