- Rating
- 3
- Confidence
- 3
- Informativeness
- 2
- Presentation
- 3
- Soundness
- 3
Read review
Paper summary
This paper proposes AUGUstus, a vision-language-action model that unifies trajectory generation and chain-of-thought (CoT) understanding for autonomous driving.
The main idea is to use CoT supervision to improve action generation without requiring CoT decoding during inference. The authors introduce AUGUstus-mode, which places action tokens before CoT tokens in a shared transformer. They also propose AUGUstus-GRPO, which jointly optimizes continuous actions and CoT through reinforcement learning.
Experiments on NAVSIMv1, NAVSIMv2, and nuScenes show strong planning performance. The paper also includes ablation studies, internal driving evaluations, and preliminary real-vehicle deployment at 12 Hz.
Critical strengths
-
A meaningful architecture that balances planning performance and inference efficiency. The paper addresses an important trade-off between using VLM understanding and maintaining low inference latency. The action-first design allows CoT supervision during training without requiring CoT generation during deployment. Table 6 shows that AUGUstus-mode achieves the same PDMS (90.6) as the CoT-first design while reducing inference time from approximately 4.6 seconds to 73.6 ms. This provides direct evidence for the practical value of the proposed ordering.
-
Strong experimental results with useful ablation studies. The method achieves competitive or state-of-the-art results across multiple public benchmarks. The authors also examine CoT content, generation order, and the effects of the two-stage GRPO training. The comparisons with the no-CoT model and the JaT-GRPO-only variant provide evidence that both CoT supervision and joint optimization contribute to performance. Multi-seed evaluations further support the stability of the reported results.
Critical weaknesses
-
The specific contribution of CoT understanding is not fully isolated. The paper shows that adding CoT supervision improves planning performance. However, the selected CoT mainly contains object bounding boxes and future-position labels. Since action tokens cannot attend to CoT tokens, the improvement comes through shared representations and training gradients. It remains unclear whether the benefit comes from CoT-based understanding or simply from additional scene prediction supervision. The existing no-CoT ablation establishes the value of additional supervision, but does not fully distinguish these explanations.
-
The benefit of the proposed unified GRPO formulation needs clearer validation. Table 7 shows that joint GRPO outperforms the variant without CoT-GRPO. This supports the benefit of optimizing both branches. However, the paper does not compare shared advantages with separate advantages for action and CoT. It also lacks a direct empirical comparison with alternative flow-based RL methods under the same training setup. These comparisons would help clarify which parts of AUGUstus-GRPO are responsible for the improvements.
-
The real-world evaluation provides limited quantitative evidence. The internal 1,000 km evaluation reports relative performance differences, but does not provide the underlying absolute metrics. The real-vehicle results mainly demonstrate inference speed and qualitative driving behavior. These results are encouraging, but they provide limited evidence about the reliability and safety of the complete system during extended real-world operation.
Questions for authors
-
Could the authors compare CoT supervision with a non-autoregressive auxiliary prediction head trained on the same bounding-box and future-position labels? This would help isolate the benefit of the proposed CoT design.
-
How does the shared-advantage GRPO objective compare with optimizing action and CoT using separate advantages under the same rewards and training budget?
-
Could the authors provide a direct comparison between JaT-GRPO and existing flow-based RL methods using the same backbone and training setup?
10 comments
Response to W1 & Q1: Isolating the Contribution of CoT Supervision (2/2)
(3) The requested control: a non-autoregressive auxiliary head on the same labels. A lightweight set-prediction head (2-layer cross-attention decoder with object queries, Hungarian matching, L1 + GIoU for boxes, cross-entropy for the future-position bins) on the context and action hidden states is trained on exactly the same objects, boxes and bins as the prediction CoT, with the language loss off and its loss weight tuned on a held-out navtrain split (Table 1, with the rows of Table 9 of the original manuscript). It reaches 89.0 PDMS, below the no-CoT baseline (89.2) and at the level of Short-prediction (89.1), vs. 90.6 for the prediction CoT: like Short-prediction, a new output space detached from the VLM's language representation disturbs the pre-trained latent space. The new shuffled and corrupted CoT targets keep the JSON format but fall well below no-CoT (87.8 and 88.1), so CoT content that does not match the scene hurts the action. The benefit of the proposed CoT design therefore comes from correct scene content expressed in the VLM's native vision-language space, not from extra labels or extra token prediction.
Table 1. CoT supervision vs. label-matched alternatives (NAVSIMv1 navtest, SFT)
| Supervision target | Labels | Encoding | PDMS↑ | NC↑ | DAC↑ | EP↑ | TTC↑ | Comf.↑ |
|---|---|---|---|---|---|---|---|---|
| No CoT (Table 9) | none | – | 89.2 | 98.8 | 96.8 | 83.5 | 95.8 | 100.0 |
| Short-prediction (Table 9) | boxes + future position | atomic vocabulary | 89.1 | 98.7 | 96.7 | 83.1 | 95.4 | 100.0 |
| Non-autoregressive auxiliary head (new) | boxes + future position | set-prediction head, no text | 89.0 | 98.9 | 96.3 | 83.1 | 95.7 | 100.0 |
| Shuffled CoT (new) | CoT of another scene, same format | JSON text | 87.8 | 98.7 | 95.6 | 81.4 | 95.7 | 100.0 |
| Corrupted CoT (new) | correct boxes, random future positions | JSON text | 88.1 | 98.5 | 95.8 | 82.6 | 95.5 | 100.0 |
| Prediction CoT (ours, Table 9) | boxes + future position | JSON text | 90.6 | 99.0 | 97.9 | 84.2 | 96.4 | 100.0 |
The language form also keeps the CoT a token sequence on the existing LM head, so it can be optimized with token-level GRPO together with the action in RL (Table 7 of the original manuscript: 92.6 without CoT-GRPO vs. 94.6 with it).
Response to W2 & Q2: Shared vs. Separate Advantages (1/2)
(1) Clarification of the advantage. We apologize for the misunderstanding regarding Sec. 3.3. In AUGUstus-GRPO the action and CoT branches share the paired rollouts, the transformer and one joint update, and their group-relative advantages are computed per branch: the action advantage from the planning reward, the CoT advantage from the CoT reward. All AUGUstus-GRPO results in the manuscript use these per-branch advantages (repository settings in (6): rl_split_advantage: true, rl_reward_norm: false). We have revised Sec. 2, Sec. 3.3 and Appendix D, Eqs. (4), (6) and (19)–(21), Fig. 3 and Algorithms 3, 4 and 7 to write the two advantages out explicitly.
(2) What AUGUstus-GRPO computes. For the paired rollouts of a context, the two rewards are normalized separately within the group:
Each advantage enters its own clipped GRPO objective (Eq. 6), and the two objectives form one loss with a single backward pass:
The advantages are separate, but the gradients are not: both terms back-propagate through the same transformer, and since the CoT tokens attend to the action tokens, the CoT-GRPO gradient reaches the action hidden states (Figure 2).
Figure 2. AUGUstus-GRPO: per-branch advantages, one joint backward pass
G paired rollouts of one context (shared initial noise x_1)
┌──────────────────────────────────┴──────────────────────────────────┐
action x_i (SDE rollout) CoT y_i (sampled after the action)
│ │
R_plan(x_i) R_pred(y_i)
Stage I: PDMS; Stage II: endpoint + safety box IoU × future-position agreement
│ normalize within the group │ normalize within the group
▼ ▼
A_i^act A_i^cot
│ │
L_act = −J_action(A^act) + β_action·KL_act L_cot = −J_CoT(A^cot) + β_CoT·KL_CoT
└──────────────────────────────────┬──────────────────────────────────┘
▼
L = L_act + α_CoT · L_cot ──▶ one backward pass
│
┌──────────────────────────────────▼──────────────────────────────────┐
│ one transformer: │ V V L L │ A A A A A A A A │ L L L L L L │ │
│ context action block CoT block │
│ CoT tokens attend to action tokens ──▶ ∂L_cot reaches the action │
│ hidden states and the shared trunk; actions never read the CoT │
└─────────────────────────────────────────────────────────────────────┘
(3) The requested comparison. We compare our per-branch advantages with a single shared advantage that drives both branches. Everything else (SFT initialization, rewards, , rollout steps, learning rate, number of updates, KL coefficients) is identical.
Table 2. Single shared advantage vs. per-branch advantages (NAVSIMv1, same rewards and training budget; Stage I + II)
| Advantage | PDMS↑ | NC↑ | DAC↑ | EP↑ | TTC↑ | Comf.↑ |
|---|---|---|---|---|---|---|
| Single shared, for both branches | 94.2 | 98.7 | 98.3 | 91.5 | 97.0 | 100.0 |
| Per-branch, and (ours) | 94.6 | 98.9 | 98.9 | 91.9 | 97.1 | 100.0 |
Thank you for your valuable and insightful comments. We deeply appreciate your input and have made every effort to address your concerns, as detailed below.
Response to W1 & Q1: Isolating the Contribution of CoT Supervision (1/2)
(1) The mechanism. In AUGUstus-mode the action tokens never attend to the CoT tokens, so the CoT acts on the action only through the shared representation and the training gradients: the CoT tokens are placed after and attend to the action tokens, so the CoT cross-entropy back-propagates directly into the action hidden states (Appendix D.2.1). CoT in AUGUstus is therefore structured supervision expressed in language, the wording the revised manuscript now uses.
(2) The manuscript already contains a same-labels, different-encoding control. Short-prediction (Table 9 of the original manuscript) encodes the same boxes and future positions as the prediction CoT in an atomic vocabulary of 2,573 new tokens (≈14 instead of ≈100 tokens per scene) and reaches 89.1, below the no-CoT baseline of 89.2, vs. 90.6 for the prediction CoT (Figure 1). The gain is therefore not "more scene-prediction supervision": it requires the supervision to be expressed in the vision-language space that the VLM is pre-trained on. The JSON
bbox_2dformat of the prediction CoT is the grounding format of the Qwen-VL pre-training, so it activates the pre-trained spatial understanding instead of training a new output space from scratch.Figure 1. Three encodings of the same per-scene annotation (risk-object boxes + future-position bins)