- Rating
- 2
- Confidence
- 4
- Informativeness
- 2
- Presentation
- 2
- Soundness
- 2
Read review
Paper summary
DyCoPart is a diffusion text-to-motion model built on a spatial-temporal Mamba backbone. Joint features from a two-layer skeleton GCN are softly routed into six part slots, using motion-conditioned logits combined with an anatomical distance prior (Dynamic Part Learning, DPL). The part tokens then go through a Part-SSM and a "directed causal coordination rectifier" (DCCR), where each part at frame t aggregates messages from the other parts at frame t−1. DCCR is trained with random part masking/freezing/noise and three regularizers on its attention matrix. Experiments report the standard HumanML3D/KIT-ML metrics plus author-defined coordination (TC/SC), physical-plausibility and part-level metrics.
Critical strengths
The same-backbone ablation (Table 4a, Table 6) is the most useful result in the paper. A hard fixed partition improves the part-level metrics but drops TC/SC from 0.603/0.742 to 0.433/0.591 and worsens all four physical metrics, while soft routing improves both sides. Provided the FP variant is a fair implementation (Q1), this turns the local-vs-global trade-off that part-based papers usually just assert into something measured. The modules are also cheap (+0.41M parameters, +5.9% latency, Table 10) and give similar gains on MDM (Table 7).
Critical weaknesses
-
Missing baselines, and gains credited to the wrong source. Table 1 leaves out the strongest recent methods, and these are exactly the joint-level / skeleton-aware models that the "holistic vs. fixed-part" framing ignores. MoGenTS (NeurIPS'24) reports FID 0.033 on HumanML3D and Top-1 0.445 / FID 0.143 on KIT-ML, all better than DyCoPart (0.041; 0.438 / 0.185). SALAD (CVPR'25) reports higher R-Precision (Top-1 0.581 vs. 0.565); it is used in the supplementary video but is neither cited nor in Table 1. LGTM, HiSTF Mamba, ParTY and BiPO, which Sec. 2.2 itself lists as the main part-based baselines, are also missing. Separately, the in-house baseline without any proposed module already reaches Top-3 0.832 (Table 4a), above every prior method in Table 1, yet Sec. 4.2 credits the +9.7% Top-1 gain over ParCo to dynamic routing and causal coordination. What the modules add is 0.832 → 0.848 Top-3 and 0.085 → 0.041 FID, reported as point estimates without seed variance.
-
Neither mechanism is shown to do what its name says. The ablation compares hard fixed parts with soft, motion-conditioned routing, so "soft" and "dynamic" are confounded. Static soft routing (prior π only) is the missing control, and time variation is shown for a single prompt (Table 9, own-slot weight 0.71 → 0.55). DCCR is claimed to recover coherent motion from corrupted or desynchronized parts (abstract, Fig. 1c), but nothing perturbs parts at inference, and neither the training perturbations nor the auxiliary losses are ablated. "Causal" amounts to a one-frame lag inside a denoiser that sees the whole sequence; "counterfactual" refers to random part masking/freezing/noise. The method description also contradicts itself in places (auxiliary losses, Part-SSM axis, part-to-joint mapping; see Q5).
Questions for authors
- How is FP implemented: hard one-hot routing into the same slots and Part-SSM? Does it replace or run alongside the baseline's spatial Mamba? The FP collapse in TC/SC carries much of the argument, so this changes how I read Table 4.
- Please add static soft routing as a DPL control, and an inference-time test that masks, freezes or noises one part slot, with TC/SC/FID reported with and without DCCR.
- Table 1 should include MoGenTS, SALAD and the cited part-based methods, plus the baseline's full row (Top-1, MM-Dist, MModality). Since all coordination and part-level metrics in Tables 2-3 are defined or adapted by the authors, a small user study on coordination would also help.
- p_t and ε_t are computed from root and foot-contact features of which signal: the noisy x_t, the predicted x_0, or something else? At high noise levels, contacts read from x_t carry no information. What is the "whole-motion branch" that produces c_t?
- Clarifications: Sec. 3.3 says the auxiliary losses target intervention response, compensatory support and output stability, while App. A defines temporal-smoothness, entropy and reciprocity penalties on A. Which is correct, and what are the weights? Part-SSM is "spatial" in the abstract and Sec. 3.1-3.2 but runs along time within each part in Eq. 8. Which is it? How are part tokens mapped back to joints?
6 comments
Q2: Static soft routing and inference-time part corruption
Overview. To address these requests, we:
- 1. Add matched prior-only and sequence-conditioned static-soft controls (Appendix C.9, manuscript Table 12; Appendix C.9, manuscript Table 12); - 2. Evaluate inference-time masking, freezing, and noise injection with and without DCCR (Appendix C.10, manuscript Table 14).
1. Matched static-soft routing controls.
Following the matched protocol described in Q1, we compare:
- Matched hard routing: fixed anatomical one-hot assignments.
- Prior-only static soft routing: the learned anatomical-prior matrix, with motion-conditioned logits removed. The prior remains learnable during training; within each trained block, its assignments are shared across samples and frames.
- Sequence-conditioned static soft routing: routing logits are averaged over valid frames, and the resulting assignment matrix is shared across the sequence.
- Dynamic soft routing: assignments are predicted separately for each motion frame.
Table 1. Routing comparison. The last four rows use matched processing; the first two are architectural references. Best in bold; second-best underlined.
| Variant | FID ↓ | TC ↑ | SC ↑ |
|---|---|---|---|
| Backbone baseline | 0.085 | ||
| Original FP | 0.433 | 0.591 | |
| Matched hard routing | 0.079 | 0.478 | 0.636 |
| Prior-only static soft routing | 0.077 | 0.531 | 0.681 |
| Sequence-conditioned static soft routing | 0.076 | 0.520 | 0.669 |
| Dynamic soft routing / DPL | 0.058 | 0.614 | 0.769 |
DPL improves FID, TC, and SC over both static-soft controls under matched processing. These results support contributions from motion-conditioned assignments and their frame dependence beyond soft allocation alone.
2. Inference-time part corruption.
We corrupt one part slot at the Part-SSM output immediately before the final DCCR block, or at the corresponding location without DCCR. Corruption is injected once, at the 16th step of the 20-step sampler, over a contiguous window covering 20% of valid motion frames.
Masking zeros the selected features; freezing repeats the state immediately preceding the window; noise injection adds Gaussian noise with per-channel standard deviation equal to half the corresponding validation feature scale. Prompts, initial sampling noise, affected slots, and windows are paired across models, with affected slots balanced across the test set. Both classifier-free guidance branches receive the same intervention.
Table 2. Inference-time corruption. Each cell reports without DCCR / with DCCR; the better value within each pair is bold.
| Condition | TC ↑ | SC ↑ | FID ↓ |
|---|---|---|---|
| Clean | 0.614 / 0.623 | 0.769 / 0.777 | 0.058 / 0.041 |
| Mask | 0.568 / 0.610 | 0.736 / 0.763 | 0.094 / 0.051 |
| Freeze | 0.547 / 0.597 | 0.717 / 0.759 | 0.109 / 0.063 |
| Noise | 0.577 / 0.604 | 0.740 / 0.768 | 0.089 / 0.058 |
The DCCR-trained model achieves better corrupted-generation scores and smaller deterioration from its own clean performance across all three metrics and corruption types. This supports robustness to the tested part-feature corruptions beyond its clean-generation advantage.
These comparisons evaluate DCCR together with its intervention training and auxiliary objectives; they do not separately isolate the effects of training perturbations and individual auxiliary losses.
We will also revise the “causal” and “counterfactual” terminology. DCCR uses directed previous-frame messages within a denoiser that processes the complete sequence, and the training interventions are random feature perturbations. The evidence supports directed coordination and corruption robustness, rather than causal identification or temporal causality of the complete denoiser.
Q3: Expanded comparisons, gain attribution, and human evaluation
Overview. To address these concerns, we:
- 1. Expand the HumanML3D and KIT-ML comparisons to include the requested methods (Appendix C.1, manuscript Tables 4–5); - 2. Separate same-backbone improvements from cross-method differences (Sec. 4.3; Appendix C.4, manuscript Table 8); - 3. Report a coordination-focused human evaluation (Appendix C.12, manuscript Table 17).
1. Expanded benchmark comparisons.
The tables add published results for joint-level, skeleton-aware, and part-based methods, which we discuss in Sec. 2.2.
Table 3. Expanded HumanML3D comparison. Baseline methods use their published results; the internal backbone and DyCoPart use our experimental results. Best in bold; second-best underlined. Diversity is reported without ranking. NR: not reported in the cited paper.
| Method | R@1 ↑ | R@2 ↑ | R@3 ↑ | FID ↓ | MM-Dist ↓ | Diversity | MModality ↑ |
|---|---|---|---|---|---|---|---|
| MoGenTS | 0.529 | 0.719 | 0.812 | 2.867 | 9.570 | NR | |
| SALAD | 0.581 | 0.769 | 0.857 | 0.076 | 2.649 | 9.696 | 1.751 |
| LGTM | 0.490 | 0.689 | 0.788 | 0.218 | 3.013 | 9.638 | NR |
| HiSTF Mamba | 0.537 | 0.724 | 0.813 | 0.039 | 2.884 | 9.689 | 1.701 |
| ParTY | 0.550 | 0.744 | 0.836 | 0.035 | 2.779 | 9.534 | 2.155 |
| BiPO | 0.523 | 0.714 | 0.809 | 0.030 | 2.880 | 9.556 | 1.374 |
| ParCo | 0.515 | 0.706 | 0.801 | 0.109 | 2.927 | 9.576 | 1.382 |
| Internal backbone | 0.547 | 0.745 | 0.832 | 0.085 | 2.774 | 9.618 | 2.561 |
| DyCoPart | 0.041 | 9.661 |
The internal backbone is the implementation used in our ablations, rather than the published Light-T2M result. HiSTF Mamba values follow arXiv v3.
Table 4. Expanded KIT-ML comparison. Reporting conventions follow Table 3; —: unavailable in this response.
| Method | R@1 ↑ | R@2 ↑ | R@3 ↑ | FID ↓ | MM-Dist ↓ | Diversity | MModality ↑ |
|---|---|---|---|---|---|---|---|
| MoGenTS | 0.445 | 0.671 | 0.797 | 0.143 | 2.711 | 10.918 | NR |
| SALAD | 0.477 | 0.711 | 0.828 | 0.296 | 2.585 | 11.097 | 1.004 |
| HiSTF Mamba | 0.678 | 0.798 | 0.217 | 2.737 | 11.160 | 1.058 | |
| ParTY | 0.449 | 2.694 | 11.210 | ||||
| BiPO | 0.444 | 0.674 | 0.803 | 0.164 | 10.833 | 1.098 | |
| ParCo | 0.430 | 0.649 | 0.772 | 0.453 | 2.820 | 10.950 | 1.245 |
| Internal backbone | 0.421 | 0.648 | 0.772 | 0.204 | 2.824 | 10.951 | — |
| DyCoPart | 0.438 | 0.667 | 0.797 | 0.185 | 2.719 | 11.073 | 0.962 |
These literature results retain each paper’s evaluation setup and are not presented as unified reruns.
SALAD reports stronger text alignment on both datasets. MoGenTS, BiPO, and ParTY report lower FID on both datasets, while HiSTF Mamba reports lower FID on HumanML3D. DyCoPart therefore does not lead across all standard metrics.
2. Attribution to the proposed modules.
The 9.7% relative R@1 improvement over ParCo is a cross-method difference and cannot be attributed entirely to DPL and DCCR. We assess their contribution against our internal backbone:
- HumanML3D: R@1 improves by 0.018, R@3 by 0.016, and FID decreases by 0.044.
- KIT-ML: R@1 improves by 0.017, R@3 by 0.025, and FID decreases by 0.019.
The same-backbone results support improvements in text alignment and generation fidelity on both datasets. MModality does not improve on HumanML3D.
3. Coordination-focused human evaluation.
We conducted a blinded pairwise study on 60 randomly selected HumanML3D test prompts, fixed before inspecting outputs. DyCoPart was compared with the internal backbone, SALAD, and ParCo. One output per prompt and method was selected using a predefined sampling rule without output screening.
Rendering and playback settings were identical; method identities were hidden, and presentation order and left/right placement were randomized. From a pool of 10 raters, five distinct raters assessed each prompt–comparison pair, yielding 300 judgments per comparison and 900 judgments. Raters selected the motion with more natural coordination among the arms, legs, and torso, with ties allowed.
Table 5. Coordination preferences. Tie-adjusted preference assigns half a vote to each tie.
| Comparison | DyCoPart wins | Ties | Comparator wins | Tie-adjusted preference |
|---|---|---|---|---|
| DyCoPart vs. internal backbone | 181 | 44 | 75 | 67.7% |
| DyCoPart vs. SALAD | 168 | 37 | 95 | 62.2% |
| DyCoPart vs. ParCo | 217 | 24 | 59 | 76.3% |
DyCoPart received higher coordination preference than every comparator on the evaluated prompts, providing perceptual evidence alongside the automatic coordination metrics.
Thank you for your careful review and valuable constructive feedback. We have carefully considered your comments and provide our detailed responses to each point below.
Q1: FP implementation and fairness of the routing comparison
Overview. To address this question, we:
- 1. Clarify the original FP implementation and its relationship to the backbone (
Appendix D.3); - 2. Define matched controls that separate routing changes from feature-processing differences (Appendix C.9, manuscript Table 12).1. Original FP implementation.
The original FP variant groups joints by fixed anatomical indices, concatenates their features, and applies part-specific projections. Its groups can overlap, and it does not use DPL’s joint encoder and skeleton GCN. Although it uses the same six slot identities and shared Part-SSM processing, it is not simply a one-hot replacement of DPL’s assignment matrix.
The part pathway runs alongside the retained whole-body pathway, and its output is fused back into the whole-body representation. Consequently, the original FP comparison differs from DPL in feature construction and aggregation as well as grouping.
The original FP degradation therefore cannot be attributed entirely to fixed grouping. We will revise the corresponding discussion to distinguish this architectural comparison from the matched routing experiments below.
2. Matched comparison protocol.
The four matched variants in Table 1 share the joint encoder and GCN, aggregation framework, Part-SSM, joint reconstruction, whole-body pathway, fusion, decoder, and training settings. Each variant is trained separately, with DCCR disabled throughout.
Only the routing rule changes: fixed anatomical one-hot assignments, prior-only static soft assignments, sequence-conditioned static soft assignments, or frame-dependent dynamic soft assignments. This comparison addresses routing under matched feature processing.