- Rating
- 3
- Confidence
- 2
- Informativeness
- 2
- Presentation
- 2
- Soundness
- 2
Read review
Paper summary
SWIM is a reinforcement-learning framework for physically simulated human swimming. Its main component represents body–water interactions using whole-body forces and torques together with per-link features encoded by a dual-branch GCN VQ-VAE. A PPO policy adjusts stroke frequency and reference-pose offsets, while a GPU-based fluid–rigid simulator supports training. Experiments cover four swimming styles and generalization to new paths, flows, fluid properties, perturbations, and body geometries.
Overall, I am slightly positive. SWIM presents a useful system for a challenging swimming-control problem, and the broad generalization experiments are a meaningful strength. My main reservation is that the necessity and advantage of the proposed tokenizer remain unclear relative to simpler force representations. Nevertheless, the system-level contribution and demonstrated capabilities are sufficient to support weak acceptance.
Critical strengths
-
The system addresses a challenging character-control problem. Swimming requires continuous full-body interaction with fluid, and the combination of reference-guided control and an efficient coupled simulator provides a practical solution. Table 3 shows substantially higher task success than the adapted imitation-learning baselines.
-
The generalization evaluation is extensive. Training uses a simple still-water task and one reference motion per style, while testing covers unseen trajectories, flows, fluid properties, external forces, and morphology changes. These results support the system’s utility beyond reproducing the reference motion.
Critical weaknesses
-
The advantage of the main environment representation is not sufficiently established. In Table 4, whole-body raw forces achieve higher success than the complete method, 69.8% versus 63.9%, with favorable position-error and roll metrics. Per-body tokens also achieve higher success. The proposed combination may offer benefits in motion quality, but the reported results do not clearly establish why it is preferable to these simpler alternatives.
-
The comparisons primarily establish system-level effectiveness. The imitation baselines are adapted to swimming, while the fluid-model comparisons replace the transition model. These comparisons do not fully isolate the contribution of the tokenizer. Moreover, high-resolution SPH and CFC receive matched training time rather than matched sample counts, limiting conclusions about their attainable control quality.
Questions for authors
-
What measurable advantage justifies combining whole-body raw forces with per-link tokens, given the stronger task metrics of the whole-body raw baseline in Table 4?
-
Does quantization measurably reduce temporal feature variation or improve training stability compared with a continuous GCN encoder?
-
Could you clarify how much of the improvement over the imitation baselines comes from the environment representation versus the reference-based action parameterization and reward design?
5 comments
Response to Reviewer miux (2/5)
1. Additional Experiments
Combined representations and EMA sensitivity. Across the tested EMA parameters, SWIM maintains the best mean SRV of 99.5-99.8%, compared with 95.5-99.3% for Whole-body raw + Per-body raw (EMA). The Whole-body raw + Per-body raw (EMA) variant usually achieves higher mean SR, while both representations have similar Vel and Accel values. In the user study at α = 0.22, matching the setting in the paper, the two methods receive the same overall mean rank.
The following table reports SRV (%) for trajectory following (T2), including interpolation (interp.) and extrapolation (extrap.) settings.
| Method | T2 | T2 interp. | T2 extrap. | T2 × E1 | T2 × E1 interp. | T2 × E1 extrap. |
|---|---|---|---|---|---|---|
| Whole-body raw + Per-body raw(EMA α = 0.22) | 100.0±0.0 | 100.0±0.0 | 100.0±0.0 | 99.2±1.3 | 99.4±0.8 | 99.0±2.1 |
| Whole-body raw + Per-body raw(EMA α = 0.2) | 98.3±3.3 | 99.3±1.4 | 96.9±6.2 | 96.1±7.4 | 96.3±6.8 | 95.8±8.3 |
| Whole-body raw + Per-body raw(EMA α = 0.1) | 97.5±5.0 | 98.6±2.8 | 95.8±8.3 | 93.0±11.8 | 94.5±9.6 | 90.6±15.0 |
| Whole-body raw + Per-body raw(EMA α = 0.05) | 97.1±4.8 | 97.9±2.7 | 95.8±8.3 | 93.4±11.4 | 93.8±10.3 | 93.0±13.0 |
| SWIM(Whole-body raw + Per-body token(EMA α = 0.22)) | 100.0±0.0 | 100.0±0.0 | 100.0±0.0 | 99.0±0.7 | 98.9±1.2 | 99.2±0.5 |
| SWIM(Whole-body raw + Per-body token(EMA α = 0.2)) | 100.0±0.0 | 100.0±0.0 | 100.0±0.0 | 99.6±0.2 | 99.5±0.3 | 99.9±0.3 |
| SWIM(Whole-body raw + Per-body token(EMA α = 0.1)) | 100.0±0.0 | 100.0±0.0 | 100.0±0.0 | 99.3±0.8 | 99.3±1.2 | 99.3±0.8 |
| SWIM(Whole-body raw + Per-body token(EMA α = 0.05)) | 100.0±0.0 | 100.0±0.0 | 100.0±0.0 | 99.3±0.6 | 99.4±0.6 | 99.2±0.7 |
All four SWIM variants achieve 100% SRV in T2 still water condition. The Whole-body raw + Per-body raw (EMA) also achieves 100% at α = 0.22, but its mean SRV decreases at the other tested values. In T2 × E1, SWIM maintains a mean SRV of 99.0-99.6%, while the Whole-body raw + Per-body raw (EMA) ranges from 93.0% to 99.2%. These results suggest that SWIM’s survival performance is less sensitive to the EMA parameter, particularly in trajectory following.
Further analysis shows that the lower SRV of the Whole-body raw + Per-body raw (EMA) representation is mainly associated with backstroke. The table below reports the backstroke results.
| Method | SRV↑(%) | SR↑(%) | PS↓(m) | LD↓(m) | Prog→1 | Roll↓(°) | Vel↓(rad/s) | Accel↓(rad/s²) |
|---|---|---|---|---|---|---|---|---|
| Whole-body raw + Per-body raw(EMA α = 0.22) | 97.6±4.2 | 63.2±7.1 | 0.207±0.047 | 0.134±0.047 | 0.891±0.018 | 2.95±3.30 | 1.24±0.12 | 57.97±6.16 |
| Whole-body raw + Per-body raw(EMA α = 0.2) | 89.3±18.4 | 61.4±4.6 | 0.216±0.047 | 0.133±0.022 | 0.882±0.021 | 5.59±7.74 | 1.19±0.03 | 56.49±4.05 |
| Whole-body raw + Per-body raw(EMA α = 0.1) | 83.1±21.0 | 56.6±5.6 | 0.233±0.049 | 0.121±0.010 | 0.867±0.020 | 8.37±10.32 | 1.12±0.02 | 52.52±2.78 |
| Whole-body raw + Per-body raw(EMA α = 0.05) | 86.6±23.0 | 62.9±6.8 | 0.232±0.056 | 0.121±0.021 | 0.859±0.019 | 6.45±8.40 | 1.16±0.04 | 55.91±0.90 |
| SWIM(Whole-body raw + Per-body token, EMA α = 0.22) | 99.7±0.4 | 58.8±4.8 | 0.225±0.045 | 0.136±0.060 | 0.870±0.008 | 1.70±1.53 | 1.24±0.08 | 56.31±6.32 |
| SWIM(Whole-body raw + Per-body token, EMA α = 0.2) | 99.7±0.4 | 58.8±8.8 | 0.252±0.105 | 0.168±0.117 | 0.875±0.010 | 1.57±0.35 | 1.23±0.07 | 56.66±4.49 |
| SWIM(Whole-body raw + Per-body token, EMA α = 0.1) | 99.8±0.2 | 57.0±2.8 | 0.226±0.008 | 0.143±0.033 | 0.873±0.017 | 1.79±0.29 | 1.17±0.07 | 54.13±4.20 |
| SWIM(Whole-body raw + Per-body token, EMA α = 0.05) | 98.7±2.0 | 64.2±7.5 | 0.198±0.035 | 0.122±0.042 | 0.884±0.020 | 1.09±0.50 | 1.14±0.05 | 54.26±2.93 |
For backstroke, SWIM maintains a mean SRV of 98.7-99.8%, compared with 83.1-97.6% for the Whole-body raw + Per-body raw (EMA) representation. SWIM also has lower SRV variability across seeds. This supports the combined tokenised representation as a more reliable choice for survival across the tested EMA settings.
Response to Reviewer miux (3/5)
1. Additional Experiments
As mentioned in Appendix C 1.1, "We do not claim that this is the optimal environment state in all metrics. Instead, our design is motivated by the trade-off between force-estimation accuracy and learning efficiency … Within the space of structured, low-dimensional representations, the proposed combination provides a practical balance between generalization performance and motion quality."
We thank the reviewer for this suggestion. Our original study did not cover all representation choices and missed Per-body raw (EMA) and Whole-body raw + Per-body raw (EMA). These additional experiments help clarify the trade-offs between the representations.
Our contribution includes a low-dimensional state representation that effectively distills complex character-fluid interactions; we also do not claim that SWIM is the best across all metrics. Having said it, based on these results of new representations, we do admit that the ones with Per-body raw (EMA) could give good results sometimes. Therefore, we are open to revising the paper to include EMA-based and tokenised variants of SWIM. In addition, we would like to mention that tokenization may also offer opportunities for future research where transformers can be used, such as combining fluid representations with language embeddings, although we have not evaluated this possibility here. We welcome the reviewer’s suggestions.
Since we cannot include videos in the rebuttal, we conducted a user study to assess perceived motion quality. The lower the score, the better.
| Stroke | SWIM | W/O environment state | Whole-body raw + Per-body raw(EMA) |
|---|---|---|---|
| All strokes | 1.92±0.35 | 2.17±0.28 | 1.92±0.30 |
| Front crawl | 2.00±0.53 | 2.25±0.53 | 1.75±0.61 |
| Backstroke | 2.08±0.69 | 1.73±0.63 | 2.19±0.66 |
| Breaststroke | 1.65±0.54 | 2.60±0.44 | 1.75±0.53 |
| Butterfly | 1.94±0.40 | 2.08±0.48 | 1.98±0.45 |
The user study collects results of 24 participants, of 91.7% reported experience in recreational or competitive swimming, coaching, or lifeguarding. Each participant evaluated 8 side view video clips covering four strokes and two goal settings per stroke (a large goal angle and a small goal angle), the same task setting in the supplementary webpage ablation study. The participant was shown a series of video clips, each of which contains the results of three methods mentioned in the table above. For each video clip, participants ranked the three methods from 1 (most natural) to 3 (least natural). The methods are unlabelled, and their order was shuffled for each question. For each stroke, all three methods use the same training seed as the supplemental material videos. The seed selected by the SWIM's best task performance instead of motion quality.
Overall, SWIM and Whole-body raw + Per-body raw (EMA) receive the same mean rank of 1.92, while W/O environment state receives a worse mean rank 2.17. This indicates W/O environment state achieves unnatural controlled motion.
Compared with Whole-body raw + Per-body raw (EMA), SWIM has a better mean rank for backstroke (2.08 versus 2.19), breaststroke (1.65 versus 1.75), and butterfly (1.94 versus 1.98), but a worse mean rank for front crawl (2.00 versus 1.75). W/O environment state receives the best mean rank for backstroke, despite being ranked least natural overall.
Response to Reviewer miux (1/5)
We thank the reviewer for their time and constructive feedback. We appreciate their positive assessment of “SWIM presents a useful system for a challenging swimming-control problem”, and their view that its “system-level contribution and demonstrated capabilities are sufficient to support weak acceptance.” Below, we present the requested additional experiment, address the reviewer’s concerns about our method, and provide the requested clarifications.
1. Additional Experiments
Thank you for this comment. We agree that these comparisons primarily evaluate the complete system. We address the tokeniser’s contribution through the additional representation ablations reported below.
We evaluated 9 additional representation settings, shown in the last nine rows below. All training and evaluation settings follow the paper ablation study. The per-body raw (EMA) inputs use the same log transformation and EMA preprocessing as the tokeniser inputs.
Without environment state. This variant achieves the highest mean SR and the best mean PS and Prog. However, it also has the highest mean joint velocity error relative to the reference motion (1.36 ± 0.24 rad/s) and relatively high mean joint acceleration (67.12 ± 13.22 rad/s²). It is ranked least natural overall in the user study below; some participants described its motion as “Actions of someone drowning”. This indicates w/o environment states leads to unnatural motion to achieve better task performance.
Per-body representations. Comparing the two Per-body representations, Per-body token has a slightly higher mean SR than Per-body raw (EMA): 68.5 ± 4.6% versus 67.1 ± 5.5%. However, Per-body raw (EMA) has lower joint angular velocity to reference motion (1.28 ± 0.22 versus 1.34 ± 0.25 rad/s) and joint acceleration (63.30 ± 13.84 versus 67.23 ± 15.13 rad/s²).