- Rating
- 2
- Confidence
- 2
- Informativeness
- 2
- Presentation
- 3
- Soundness
- 2
Read review
Paper summary
This paper proposes Dense-RAM, a reinforcement-learning post-training method for a flow-based UV seam generator. It combines chart- and layout-level UV feedback into edge-aligned scores, then distributes each candidate’s advantage across mesh edges while preserving its mean. The edge-level coefficients modify RAM’s velocity regression target. Paired experiments with RAM, AWM, and DiffusionNFT report better UV topology, less fragmentation, and higher atlas utilization than scalar feedback.
Critical strengths
- The edge-level credit allocation is a novel, well-motivated extension of scalar RAM. It uses structured UV feedback while preserving each candidate’s mean advantage, without adding trainable parameters.
- The paper is clearly written and well argued. Its theoretical analysis, matched scalar–dense comparisons, and reward ablations provide substantial support for the proposed method.
Critical weaknesses
- The edge scores reflect chart-level outcomes, not counterfactual seam decisions. Edges within the same distorted chart may receive indistinguishable feedback, limiting the method’s ability to identify where a missing seam should be added.
- Preserving the mean advantage does not preserve the update’s magnitude or direction. The paper lacks shuffled-weight or update-matched controls, multi-seed results, and paired distortion and failure-rate measurements.
Questions for authors
- For a high-distortion chart that would benefit from an additional seam, how does the method distinguish edges whose cutting would reduce distortion from other uncut edges in the same chart? Could you provide a targeted analysis?
- Could the authors report results on other publicly available test sets or established UV benchmarks to assess generalization beyond the current Objaverse split?
4 comments
W2. Magnitude and direction; missing controls. Mean preservation alone does not fix the magnitude or direction of the update, so we tested both directly. None of the controls accounts for the gain.
- Random weights (the shuffled-weight control). We keep each candidate's weight values but assign them to its edges uniformly at random. The weights are then independent of local UV quality, while their mean, second moment and clipped fraction equal those of Dense-RAM exactly. All-disk falls from 64.98% to 23.49%.
- Update-matched controls. With the within-candidate standard deviation of , , so the RMS coefficient of Dense-RAM is at most times that of scalar RAM. Reward coefficient , , : 39.72%, 40.17%, 38.24%. Learning rate , : 42.19%, 41.36%.
- Direction. On identical batches at steps 0/100/250/500, the dense correction has a component [0.02, 0.06] along the scalar reward gradient and an orthogonal component [0.52, 0.64]. It is almost entirely a change of direction. Training with the orthogonal component alone recovers 89% of the gain; with the parallel component alone it matches scalar RAM.
- Seeds. Three training seeds for RAM (Table B): all-disk 40.6 ± 1.9 vs. 64.7 ± 2.1, islands 54.5 ± 1.8 vs. 37.5 ± 1.7.
- Paired statistics. Bootstrap over the 500 test meshes: +24.5 points all-disk (95% CI [21.0, 28.0]). Island counts are heavy-tailed, so we report the mean difference −17.4 (CI [−24.6, −10.5]) and the median difference −9 (CI [−12, −7]). Dense is better in all-disk on 41% of meshes, tied on 49%, worse on 10%.
- Distortion. Dense allocation lowers angular and area distortion under all three objectives.
Table B. Controls, distortion and seed variation (500 steps). Visibility and utilization are as in Table 3.
| Objective | Variant | All-disk (%) ↑ | Angle ↓ | Area ↓ | Islands ↓ |
|---|---|---|---|---|---|
| — | SeamFlow | 21.94 | 0.13 | 0.16 | 66.42 |
| DiffusionNFT | Scalar | 47.98 | 0.121 | 0.140 | 49.78 |
| DiffusionNFT | Dense | 59.58 | 0.114 | 0.129 | 42.18 |
| AWM | Scalar | 44.24 | 0.123 | 0.144 | 52.20 |
| AWM | Dense | 56.18 | 0.116 | 0.132 | 44.38 |
| RAM | Scalar () | 39.72 | 0.125 | 0.146 | 55.1 |
| RAM | Scalar () | 38.24 | 0.130 | 0.150 | 56.09 |
| RAM | Random weights | 23.49 | 0.130 | 0.160 | 65.3 |
| RAM, 3 seeds | Scalar | 40.6 ± 1.9 | 0.125 ± 0.004 | 0.145 ± 0.005 | 54.5 ± 1.8 |
| RAM, 3 seeds | Dense | 64.7 ± 2.1 | 0.111 ± 0.003 | 0.121 ± 0.004 | 37.5 ± 1.7 |
Every control you requested is thus in place, and each supports the same conclusion: the gain comes from aligned allocation. The revised version will add these controls, seed variation, distortion, the decomposition measurements, the localization analysis with two figures, and the Toys4K evaluation.
Dear Reviewer isL4,
We would appreciate your feedback on our rebuttal. Please let us know if any concerns remain. Thank you for your time and consideration.
The rebuttal substantially strengthens the empirical evidence and addresses most of my experimental concerns. My remaining reservation concerns the significance of the methodological contribution: the approach redistributes candidate-level feedback using chart-level quality, rather than distinguishing the contributions of individual edges within a chart. Although the localization analysis and additional evaluations demonstrate practical benefits, I remain unconvinced that this extension provides a sufficiently substantial advance in credit assignment. I therefore retain my weak-rejection recommendation, while acknowledging the authors’ substantial efforts and the improved experimental support.
Thank you for the positive assessment and the concrete requests. We ran every control you listed, and each one supports the allocation: with random weights of identical magnitude the all-disk rate falls from 64.98% to 23.49%, and scalar updates of matched or larger magnitude stay at 38.24–42.19%. All numbers are measured on the 500 held-out test meshes, and all tables will be added to the revised version.
W1 / Q1. Edges inside one high-distortion chart. Within a single candidate they are not distinguished, and the method does not rely on it: every uncut edge interior to the chart receives the same coefficient. Dense-RAM is edge-resolution allocation of candidate-level credit using region-level feedback, not per-edge attribution, and we will state this in Section 4.1.
The distinction comes from on-policy variation across the G=24 candidates of a mesh. Candidates that leave the chart intact have lower bg, and that region carries their largest negative coefficients. Candidates that cut it at a useful location produce passing sub-charts, obtain higher bg, and those sub-charts carry their largest positive coefficients. A cut at an unhelpful location leaves a failing sub-chart and is not reinforced. Which edge to cut is thus learned from which sampled cuts succeed, and Appendix A.2 shows that all these cases push in the same parameter direction.
Targeted analysis (Table A). We sample SeamFlow, scalar RAM and Dense-RAM from identical initial noise and compare region by region. In charts that fail the distortion test under SeamFlow, 40.3% of the seams added by Dense-RAM fall in the top quartile of per-face distortion, against 31.0% for scalar RAM and 25% for a random edge; 55% of these charts are repaired, against 30%. Over the whole test set Dense-RAM repairs 60.5% of the failing area and damages 4.2% of the passing area; scalar RAM repairs 35.2% and damages 8.1%. A regional signal is therefore enough: the policy places new seams where distortion is concentrated, with less collateral change to valid regions.
Table A. Localization (paired sampling). Repair: failing under SeamFlow, passing after. Damage: the converse.
Q2. Other test sets. We evaluated the same checkpoints, without further training, on Toys4K. The ordering is preserved: Dense-RAM improves all-disk over scalar RAM by 20.66 percentage points and reduces islands by 10.56, with lower distortion.