- Rating
- 3(+1)
- Confidence
- 3
- Informativeness
- 2
- Presentation
- 3
- Soundness
- 2
Read review
Paper summary
FieldDrive applies the Drifting paradigm to one-step trajectory generation for end-to-end autonomous driving. During imitation-learning pretraining, the method constructs a scene-conditioned reference distribution from the expert trajectory and high-PDMS trajectories selected from a clustered vocabulary. During reinforcement-learning fine-tuning, it combines the Drift Field with trajectory gradients from an online critic to improve planning performance while retaining a reference-behavior constraint. At inference, the direct variant produces a single trajectory using zero noise and one generator forward pass. A separately fine-tuned variant supports candidate generation and scorer-based selection. The paper reports competitive planning results and efficient decoding on NAVSIM.
Critical strengths
-
An interesting application of Drifting to autonomous driving. Adapting this recent generative modeling paradigm to end-to-end planning is a worthwhile direction. The combination of distribution matching and reward-guided refinement forms a coherent framework for training a one-step planner.
-
Strong planning performance with efficient one-step decoding. The direct variant achieves 92.1 PDMS without candidate scoring at inference, requiring only one generator forward pass at a reported decoding speed of 546 FPS. The scorer-assisted variant further reaches 93.3 PDMS. The positive-sample and field-component ablations provide useful evidence for the training design.
Critical weaknesses
-
The overall framework is heavily tailored to PDMS. PDMS influences both stages of training: it is used to select positive trajectories during imitation-learning pretraining, while the critic supervision and reward potential are constructed around its component metrics during RL fine-tuning. Although these choices produce strong benchmark results, they leave uncertainty about whether the improvements reflect broader driving quality. Moreover, the v1 and v2 results use different fine-tuning configurations, so the paper does not establish how the same inferred trajectories perform under both PDMS and EPDMS.
-
The effectiveness of the framework under broader and more complex reward combinations remains insufficiently explored. According to Appendix B.3, the v2 adaptation adds DDC guidance and reduces the EP weight, but does not report explicitly incorporating TLC, LK into the reward optimization. This is relevant to Table 2: despite achieving the highest aggregate EPDMS, FieldDrive has substantially lower EC, lower LK and TLC than most listed methods. These results motivate examining whether the reward-field framework can effectively optimize a more comprehensive set of driving objectives and achieve a better balance across them.
Questions for authors
-
Could the authors report both PDMS and EPDMS for the same fixed inference trajectories from each variant, particularly the EPDMS of the model achieving 92.1 PDMS? Please clarify any differences between the evaluation protocols.
-
Could the authors provide the complete v2 reward formula and explore a broader reward combination that explicitly includes additional objectives such as TLC and LK? Reporting the full metric breakdown before and after fine-tuning, together with an analysis of the low EC, would help establish whether the framework can address these weaknesses while retaining its planning performance.
-
How sensitive is performance to the nonzero α and β weights? Table 5 establishes the benefit of combining the two fields, but a controlled sensitivity analysis would help assess the robustness of their balance.
8 comments
Response (Part 2)
W2/Q2: Exploring EPDMS as a Training Reward.
For the NAVSIM v2 variant, the complete reward is
In implementation, we compute the Reward Field Vector by differentiating the logarithmic reward constructed from the critic's metric predictions.
Adding TLC and LK improves EPDMS.
We explored adding TLC and LK to the reward, extending the variant with adjusted EP and DDC weights. With these objectives included and a larger α to keep trajectories closer to the expert reference, EPDMS improves from 89.71 to 90.40. TLC increases from 99.63 to 99.79, and LK increases from 94.37 to 97.79. The stronger expert constraint also encourages a more consistent trajectory style, which we associate with the improvement in EC from 73.72 to 84.30. Specifically, we set α = 2.5 in this experiment: the additional reward terms contribute gradients that are summed in the Reward Field Vector, so we increase α to balance their combined influence with the expert constraint. However, NC, DAC, and HC decrease. Table W4 reports the full breakdown, showing that this reward configuration improves overall performance while introducing trade-offs among individual objectives.
Table W4. EPDMS evaluation with additional reward objectives.
| Reward configuration | NC | DAC | EP | TTC | DDC | HC | EC | LK | TLC | EPDMS |
|---|---|---|---|---|---|---|---|---|---|---|
| Main PDMS-based reward | 97.85 | 98.63 | 93.73 | 97.09 | 97.68 | 98.03 | 57.62 | 92.52 | 99.42 | 86.63 |
| PDMS-based reward with adjusted EP and DDC weights | 98.55 | 99.05 | 88.70 | 98.04 | 99.67 | 98.18 | 73.72 | 94.37 | 99.63 | 89.71 |
| + TLC and LK | 98.04 | 98.13 | 90.05 | 97.43 | 99.57 | 97.86 | 84.30 | 97.79 | 99.79 | 90.40 |
| + TLC, LK, EC, and HC, with historical conditioning for the critic | 97.32 | 95.75 | 89.30 | 95.89 | 99.58 | 97.29 | 68.21 | 96.19 | 99.83 | 85.24 |
Adding EC and HC degrades performance under misaligned conditioning.
We further incorporated EC and HC and repeated both training stages. We tried different α values, but none yielded satisfactory performance. The configuration reported in Table W4 achieves 85.24 EPDMS, compared with 90.40 for the TLC-and-LK configuration. EC requires a historical model prediction, while HC requires ego states from the preceding three frames. To obtain EC labels, we used a frozen copy of the current model to generate historical trajectories, paired them with current-frame trajectories generated from the same initial noise samples, and computed EC using the official NAVSIM v2.2 procedure. For the critic, we encoded the historical prediction for EC and the preceding three ego states for HC, then concatenated each encoding with the current trajectory token before its corresponding scoring head.
Response (Part 3)
Adding EC and HC degrades performance under misaligned conditioning.
We further incorporated EC and HC and repeated both training stages. We tried different α values, but none yielded satisfactory performance. The configuration reported in Table W4 achieves 85.24 EPDMS, compared with 90.40 for the TLC-and-LK configuration. EC requires a historical model prediction, while HC requires ego states from the preceding three frames. To obtain EC labels, we used a frozen copy of the current model to generate historical trajectories, paired them with current-frame trajectories generated from the same initial noise samples, and computed EC using the official NAVSIM v2.2 procedure. For the critic, we encoded the historical prediction for EC and the preceding three ego states for HC, then concatenated each encoding with the current trajectory token before its corresponding scoring head.
Table W5 highlights the resulting actor–critic conditioning mismatch: the critic receives the historical information needed to predict EC and HC, while the actor retains its current-scene inputs. In particular, conditioning the actor on its own historical prediction introduces a recursive dependency. Generating the prediction for the previous frame would require an even earlier prediction, which in turn would require another preceding prediction. Our current training setup does not provide a rollout and initialization mechanism to resolve this chain, so we did not add these historical conditions to the actor.
We consider this conditioning mismatch a likely cause of the poorer result. The critic supplies reward gradients that depend on a particular historical prediction or ego-state history, but the actor cannot associate these directions with the historical conditions it does not observe. Consequently, the additional gradients, particularly those from EC, may act partly as noise during actor optimization and interfere with learning.
Table W5. Metric-wise conditioning availability for the Actor Net and Critic Net.
Current-frame conditions comprise scene features, current ego state, and the navigation command. Historical-frame conditions comprise the historical model prediction for EC and ego states from the preceding three frames for HC. ✓: the required conditioning is available and the metric is optimized by the actor or predicted by the critic. ✗: the metric is required, but its historical conditioning is unavailable. Blank: the metric is not used by that network's objective. In the main model, DDC supervises the critic but has zero reward weight for the actor.
M1: Main PDMS-based model; M2: Adjusted EP + DDC; M3: + TLC + LK; M4: + TLC + LK + EC + HC. Actor and Critic denote Actor Net and Critic Net, respectively.
| Metric | M1 Actor | M1 Critic | M2 Actor | M2 Critic | M3 Actor | M3 Critic | M4 Actor | M4 Critic | Conditioning information |
|---|---|---|---|---|---|---|---|---|---|
| NC | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | Current-frame information |
| DAC | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | 〃 |
| DDC | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | 〃 | |
| TTC | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | 〃 |
| EP | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | 〃 |
| C | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | 〃 |
| TLC | ✓ | ✓ | ✓ | ✓ | 〃 | ||||
| LK | ✓ | ✓ | ✓ | ✓ | 〃 | ||||
| EC | ✗ | ✓ | Historical-frame information | ||||||
| HC | ✗ | ✓ | 〃 |
〃 denotes the same conditioning category as the preceding row.
Our reward setting aligns with mainstream NAVSIM RL methods.
Our PDMS-based reward follows the setting used by most existing RL post-training methods on NAVSIM like ReCogDrive [2], SimVTA [3], and DriveFine [4]. We consider the historical dependence of some EPDMS metrics a plausible reason for this preference: PDMS objectives can be evaluated against the current scene and candidate trajectory, whereas optimizing EC and HC additionally requires historical conditions that may be unavailable to a planner conditioned on the current scene. For our current architecture, PDMS therefore provides a training reward better aligned with the actor's available information, while EPDMS provides a broader evaluation of driving quality.
Response (Part 1)
We appreciate your careful examination of the reward design and evaluation and present cross-evaluation results, broader reward experiments, and an analysis of the field-weight sensitivity.
Cross-evaluation under PDMS and EPDMS.
We agree that reward hacking is a potential concern during RL post-training [1]. Designing rewards that lead to better performance across a broad range of driving tasks remains an open problem. According to your suggestion, we evaluate both the main PDMS-oriented model and the EPDMS-oriented variant under both scoring protocols on the same navtest split. For each variant, the same inference trajectories are evaluated under PDMS and EPDMS. The main model achieves 92.14 PDMS and 86.63 EPDMS, while the EPDMS-oriented variant achieves 91.51 PDMS and 89.71 EPDMS. In our manuscript, we use the same PDM evaluator and metric definitions as in navtest to compute rewards on training scenes like ReCogDrive [2], SimVTA [3], and DriveFine [4].
Table W1 reports the metric breakdown under PDMS evaluation.
Table W1. Cross-evaluation of both model variants under PDMS.
Table W2 reports the metric breakdown under EPDMS evaluation.
Table W2. Cross-evaluation of both model variants under EPDMS.
Zero-shot generalization to closed-loop HUGSIM evaluation.
We additionally evaluate transfer beyond NAVSIM using the closed-loop HUGSIM benchmark. Because some HUGSIM evaluation datasets do not support rear-camera input, we retrain a version using the current front, left, and right views together with the front view from the previous frame (0.5 seconds earlier). All other configurations remain unchanged with EPDMS-oriented variant. This version is trained on the original NAVSIM training data and evaluated on HUGSIM without HUGSIM training or fine-tuning. Table W3 reports the closed-loop results.
Table W3. Zero-shot closed-loop evaluation on HUGSIM.