- Rating
- 3(+1)
- Confidence
- 2
- Informativeness
- 2
- Presentation
- 2
- Soundness
- 3
Read review
Paper summary
This paper introduces OneVLA, which unifies navigation and manipulation within a single framework. It proposes a unified action output head and a multi-stage training strategy intended to promote mutual improvement between these two tasks. The paper evaluates the framework on navigation and manipulation benchmarks, including mobile manipulation, and presents real-world experiments and ablation studies.
Critical strengths
-
Unifying navigation and manipulation is a meaningful research direction
OneVLA addresses both tasks within a shared architecture, and the reported improvements from joint training suggest that learning these capabilities together can be beneficial.
-
The evaluation covers both simulation and real-world settings
The additional mobile manipulation experiments on RoboCasa365 are particularly relevant, as they extend the validation beyond evaluating navigation and manipulation separately.
Critical weaknesses
-
The baseline comparisons are not sufficiently strong for broad state-of-the-art claim
For example, OpenVLA-OFT is included in the main comparison but omitted from the LIBERO results, where the paper instead emphasizes improvements over the original OpenVLA. Including stronger baselines under comparable settings would help establish whether the reported gains remain substantial.
-
The contribution of the unified action head is not clearly isolated
Although the paper includes joint-training ablations and a UniVLA variant with the proposed head, these do not directly compare a unified head against separate navigation and manipulation heads using the same backbone and training recipe. This makes it difficult to determine whether the improvements come from the proposed architectural design or from joint training and the multi-stage training strategy.
Questions for authors
-
Baseline clarification
Why is OpenVLA-OFT included in the main comparison but omitted from the LIBERO evaluation? Are there differences in evaluation settings that prevent a fair comparison, and does OneVLA retain its advantage under comparable settings?
-
Evidence for the architectural contribution
Have you compared the unified action head with separate navigation and manipulation heads using the same backbone, training data, and training strategy? Such results would help determine whether the unified head itself contributes to the improvements beyond joint training.
-
Navigation evaluation clarification
The paper reports OSR as the main navigation metric. Does OneVLA also outperform the baselines in SR and SPL, which account for successful stopping and path efficiency?
6 comments
Thanks and Response to Reviewer 6Yr2 (2/2)
Q3: Query about navigation metrics on VLN-CE.
Thanks for your question. OneVLA achieves higher SR and SPL than the listed baselines on both R2R-CE and RxR-CE.
(A) SR and SPL comparison. We measured NE, SR, and SPL on the R2R-CE and RxR-CE val-unseen splits and nDTW on RxR-CE. Tables 17 and 18 show these results alongside the original OSR and the baseline scores. The sources below each table identify the published results and the StreamVLN version.
Rebuttal Table 17. Expanded R2R-CE metrics. OneVLA's OSR is from original Table 1.
| Method | Params | Unified | NE ↓ | OSR ↑ | SR ↑ | SPL ↑ |
|---|---|---|---|---|---|---|
| NaVILA | 7B | ✗ | 5.22 | 62.5 | 54.0 | 49.0 |
| StreamVLN [ICRA’26] | 7B | ✗ | 4.98 | 64.2 | 56.9 | 51.9 |
| UniVLA-Navi.* (OneVLA baseline) | 7B | ✗ | 6.71 | 47.1 | 38.2 | 34.5 |
| OneVLA (Ours) | 3B | ✓ | 4.72 | 68.6 | 58.4 | 53.1 |
Rebuttal Table 18. Expanded RxR-CE metrics. OSR scores are from original Table 1.
| Method | Params | Unified | NE ↓ | OSR ↑ | SR ↑ | SPL ↑ | nDTW ↑ |
|---|---|---|---|---|---|---|---|
| NaVILA | 7B | ✗ | 6.77 | 53.6 | 49.3 | 44.0 | 58.8 |
| StreamVLN [ICRA’26] | 7B | ✗ | 6.22 | 55.7 | 52.9 | 46.0 | 61.9 |
| UniVLA-Navi.* (OneVLA baseline) | 7B | ✗ | 8.35 | 26.3 | 21.4 | 18.2 | 37.9 |
| OneVLA (Ours) | 3B | ✓ | 5.98 | 58.2 | 53.7 | 46.8 | 62.6 |
OneVLA leads StreamVLN v1 by 1.5 points in SR and 1.2 in SPL on R2R, and by 0.8 points on both metrics on RxR.
(B) How stopping is learned. We give the rare STOP command a loss weight of 15 and test this choice in original Table 7. The added SR and SPL results show how often the model stops successfully and how efficiently it completes each route. We will add these metrics and clarify the STOP weighting in the revision.
We hope these responses address your concerns, and we would be glad to answer any further questions.
To Reviewer 6Yr2
Dear Reviewer 6Yr2,
We hope this message finds you well. With the deadline of discussion period is approaching, we wanted to make sure we have fully addressed your concerns. If you have any further points or feedback you would like us to consider, please let us know. Your insights are invaluable to us, and we are open to any further suggestions you might have to improve the paper.
Thank you for the time and effort you have dedicated to reviewing our paper.
Best regards,
The Authors
Thanks and Response to Reviewer 6Yr2 (1/2)
We sincerely thank the reviewer for the thoughtful and constructive feedback. We greatly appreciate your recognition of our “meaningful research direction” and “particularly relevant RoboCasa365 experiments”. Below, we address your main questions and concerns in detail.
Thanks for your question. We included OpenVLA-OFT in SimplerEnv. The original LIBERO table focused on a cross-task comparison with UniVLA. Table 14 adds the published LIBERO results for OpenVLA-OFT, π0, and π0-FAST.
Rebuttal Table 14. Expanded LIBERO baseline comparison. OneVLA's scores reproduce original Table 5. External scores come from the papers linked below.
Evaluation settings. OneVLA achieved 97.7% average success and 96.5% on LIBERO-Long. OpenVLA-OFT reports 97.1% and 94.5%, respectively, with a separate 7B policy for each suite using third-person and wrist images plus proprioceptive state. OneVLA uses one 3B policy that also supports navigation. OpenVLA-OFT also reports 96.8% with a single policy across all four suites in its Table XIV, 0.9 points below OneVLA. We will include the stronger baselines and clarify their inputs and policy settings in the revision.
Thanks for your question. We compared a shared DiT action head with separate navigation and manipulation heads in both the single-stage and full-training settings. In each comparison we kept the backbone, data, and training recipe fixed.
(A) Same backbone and single-stage training. We used the setup from original Table 2b and added two controls, one with two full DiT-B heads and one with two heads whose combined parameter count matches the unified head. All joint models used Qwen2.5-VL-3B, the same data, a 6:4 task sampling ratio, and the same number of training steps. Table 15 shows the results.
Rebuttal Table 15. Separate-head controls extending original Table 2b. Single-task and unified-head scores are existing results. Both separate-head rows are new.
Joint training with separate heads shows how much the tasks gain from sharing the backbone. Compared with two full heads, the unified head adds 2.7, 3.3, and 2.7 points on R2R, RxR, and SimplerEnv. The parameter-matched comparison checks whether the gains hold at the same total head size.
(B) Same full training recipe. We also trained the separate-head and unified-head models on the same three-stage, 60K-step schedule. Table 16 shows the results.
Rebuttal Table 16. Head comparison under the complete training strategy. OneVLA's scores are from the original paper. The separate-head row is a new control.
The unified head improves R2R, RxR, and SimplerEnv by 3.6, 3.9, and 3.6 points, respectively. Original Table 8 reports a complementary experiment in which we transferred the unified head to UniVLA. We will add both controlled comparisons to the revision and explain what they show about the unified head.