- Rating
- 3
- Confidence
- 3
- Informativeness
- 2
- Presentation
- 3
- Soundness
- 2
Read review
Paper summary
LightDraft accelerates the pretrained Lumina-DiMOO image editor without further training. It decodes complete intermediate token predictions into look-ahead images, compares them with a reconstructed source to guide two-stage regional feature caching, and sets a bounded sampling budget from changes between these predictions. The implementation also applies region-guided residual correction to the decoded output. Experiments on 637 ImgEdit instructions and 700 PIE-Bench tasks report roughly eightfold speedups over 64-step inference with improved aggregate editing and background-preservation scores.
Critical strengths
-
The method makes useful use of intermediate predictions that are already available before remasking. Comparing decoded predictions with the reconstructed reference, aggregating early candidate regions, and performing a full refresh before refining the support are sensible ways to manage uncertainty in early localization. The paper also explicitly recognizes that cached features can become stale under bidirectional attention and includes decoding and comparison costs in end-to-end latency.
-
The experimental results look great. On ImgEdit, Table 1 reduces latency from 49.19 to 6.04 seconds while increasing overall EditScore from 4.520 to 4.786. On PIE-Bench, Table 2 reports 6.02 seconds, background PSNR of 31.28 versus 26.08, and CLIP-Edit of 21.49 versus 21.37.
Critical weaknesses
-
The preservation improvement cannot be attributed to the proposed caching mechanism from the reported ablation. Section 3.3 adds the reference reconstruction residual outside the expanded edit support directly to the final RGB output. Table 3 groups two-stage caching, minimum-support expansion, and this residual correction together as R. R alone already obtains PSNR 31.70 and LPIPS 0.029, compared with 31.28 and 0.031 for the complete method. A correction-only control, caching without correction, and the same correction applied to a comparable baseline are needed to identify the source of the headline preservation result.
-
The main uniform-reduction control uses 24 steps, whereas the adaptive rule is constrained to 12 to 18. The adaptive-only row improves latency from 18.36 to 10.74 seconds with almost unchanged CLIP-Edit, but this does not show that allocating steps according to prediction changes is better than simply choosing a fixed shorter schedule. Figure 2 explicitly contains schematic rather than measured distributions, and no distribution of allocated budgets or comparison against fixed 12-, 15-, and 18-step schedules is reported.
Questions for authors
How much of the background gain survives when residual correction is disabled, and how much can the same correction recover for uniform-step inference or an existing cache baseline using the same estimated support? This is the most consequential missing control.
At matched mean latency, how does adaptive allocation compare with a fixed budget or with the same budget distribution randomly reassigned across requests? Please report the allocation distribution and whether larger budgets correlate with actual quality improvements, rather than only with prediction change.
6 comments
Q3.2. The main uniform-reduction control uses 24 steps, whereas the adaptive rule is constrained to 12 to 18. The adaptive-only row improves latency from 18.36 to 10.74 seconds with almost unchanged CLIP-Edit, but this does not show that allocating steps according to prediction changes is better than simply choosing a fixed shorter schedule. Figure 2 explicitly contains schematic rather than measured distributions, and no distribution of allocated budgets or comparison against fixed 12-, 15-, and 18-step schedules is reported.
Thank you for this suggestion. We agree that the 24-step control alone does not adequately address comparisons with shorter fixed schedules. We therefore add original Lumina-DiMOO inference with 12, 15, and 18 sampling steps in Rebuttal Table SM-2, together with 14-step inference for comparison with LightDraft’s measured average of 14.15 steps per image.
Even without RRC, LightDraft achieves an EditScore comparable to original 18- and 24-step inference. Compared with original 15-step inference, it improves EditScore from 4.402 to 4.552 while reducing latency from 11.27 to 6.13 s/image (1.84× speedup). Against the closer 14-step control, it also improves EditScore from 4.396 to 4.552 with 1.72× acceleration. Feature reuse reduces computation per step, allowing LightDraft to run faster even than original 12-step inference (6.13 vs. 9.02 s/image).
Rebuttal Table SM-2: Additional comparisons with fixed sampling schedules on ImgEdit. The Speedup column is relative to original 64-step Lumina-DiMOO inference (49.19 s/image).
| Method | Mean steps | EditScore↑ | Following↑ | Consistency↑ | Perceptual↑ | Latency (s/image)↓ | Speedup↑ |
|---|---|---|---|---|---|---|---|
| Lumina-DiMOO, 64 steps (paper baseline) | 64 | 4.520 | 5.641 | 6.968 | 5.349 | 49.19 | 1.00× |
| Lumina-DiMOO, 24 steps | 24 | 4.529 | 5.790 | 6.920 | 5.090 | 18.06 | 2.72× |
| Lumina-DiMOO, 18 steps | 18 | 4.493 | 5.796 | 6.921 | 5.088 | 13.52 | 3.64× |
| Lumina-DiMOO, 15 steps | 15 | 4.402 | 5.690 | 6.765 | 4.993 | 11.27 | 4.36× |
| Lumina-DiMOO, 14 steps | 14 | 4.396 | 5.699 | 6.798 | 4.992 | 10.52 | 4.68× |
| Lumina-DiMOO, 12 steps | 12 | 4.342 | 5.607 | 6.761 | 4.956 | 9.02 | 5.45× |
| LightDraft w/o RRC | 14.15 | 4.552 | 5.780 | 6.795 | 5.132 | 6.13 | 8.02× |
| LightDraft | 14.15 | 4.957 | 5.941 | 7.351 | 5.869 | 6.13 | 8.02× |
We also report the empirical distribution of allocated sampling budgets across all 637 ImgEdit samples in Rebuttal Table SM-3. The measured mean is 14.15 steps, with individual budgets ranging from 12 to 18 steps.
Rebuttal Table SM-3: Measured sampling-budget distribution on ImgEdit.
| Allocated steps | Number of samples | Percentage |
|---|---|---|
| 12 | 117 | 18.37% |
| 13 | 157 | 24.65% |
| 14 | 122 | 19.15% |
| 15 | 109 | 17.11% |
| 16 | 60 | 9.42% |
| 17 | 33 | 5.18% |
| 18 | 39 | 6.12% |
| Total | 637 | 100.00% |
Regarding Figure 2, the illustrated distributional trends are consistent with our measurements. We will replace the schematic plots with plots based on measured data.
Q3.3. How much of the background gain survives when residual correction is disabled, and how much can the same correction recover for uniform-step inference or an existing cache baseline using the same estimated support? This is the most consequential missing control.
Thank you for this suggestion. Rebuttal Table SM-4 reports LightDraft with and without RRC, together with original Lumina-DiMOO inference and a 14-step baseline with RRC.
On PIE-Bench, without RRC, LightDraft preserves the background comparably to original 18- and 24-step inference, with background MSE of 0.005170 versus 0.005116 and 0.005134, respectively. It also achieves lower MSE than original 14-step inference (0.005437). Enabling RRC further reduces LightDraft’s background MSE by 46.0%, from 0.005170 to 0.002791, demonstrating its substantial contribution to background preservation.
For uniform-step inference, applying RRC to the original 14-step model improves ImgEdit EditScore from 4.396 to 4.708 and consistency from 6.798 to 7.282. With RRC enabled in both methods, LightDraft achieves an EditScore of 4.957 while remaining 1.72× faster (6.13 vs. 10.52 s/image). These results demonstrate that RRC also benefits uniform-step inference on ImgEdit; the corresponding PIE-Bench background recovery remains unquantified.
Rebuttal Table SM-4: RRC ablations and uniform-step comparisons. Background MSE is measured on PIE-Bench; all other metrics are measured on ImgEdit. The Speedup column is relative to original 64-step Lumina-DiMOO inference on ImgEdit (49.19 s/image).
| Method | Mean steps | EditScore↑ | Following↑ | Consistency↑ | Perceptual↑ | Latency (s/image)↓ | Speedup↑ | Background MSE↓ |
|---|---|---|---|---|---|---|---|---|
| Original Lumina-DiMOO, 14 steps | 14 | 4.396 | 5.699 | 6.798 | 4.992 | 10.52 | 4.68× | 0.00543709 |
| Lumina-DiMOO, 14 steps + RRC | 14 | 4.708 | 5.792 | 7.282 | 5.824 | 10.52 | 4.68× | — |
| Original Lumina-DiMOO, 18 steps | 18 | 4.493 | 5.796 | 6.921 | 5.088 | 13.52 | 3.64× | 0.00511588 |
| Original Lumina-DiMOO, 24 steps | 24 | 4.529 | 5.790 | 6.920 | 5.090 | 18.06 | 2.72× | 0.00513444 |
| LightDraft w/o RRC | 14.15 | 4.552 | 5.780 | 6.795 | 5.132 | 6.13 | 8.02× | 0.00517001 |
| LightDraft | 14.15 | 4.957 | 5.941 | 7.351 | 5.869 | 6.13 | 8.02× | 0.00279071 |
Thank you for this suggestion. We agree that grouping regional caching, support expansion, and RRC in Table 3 makes their individual contributions unclear. We therefore added separate component ablations and a 14-step Lumina-DiMOO baseline with RRC in Rebuttal Table SM-1 below.
RRC substantially improves background preservation. On PIE-Bench, adding RRC to LightDraft reduces background MSE from 0.005170 to 0.002791, a 46.0% reduction. RRC uses the expanded edit support estimated from look-ahead images to apply the reference reconstruction residual to the estimated background. This region-guided correction is itself part of LightDraft’s contribution. However, even without RRC, LightDraft achieves background preservation comparable to the original model using 24 or 18 steps (0.005170 vs. 0.005134 and 0.005116), and better than the 14-step baseline (0.005437). Support expansion provides a smaller additional benefit: with RRC disabled, adding support expansion reduces MSE from 0.005283 to 0.005170.
On ImgEdit, LightDraft without RRC also achieves an EditScore comparable to original 24- and 18-step inference. Using an average of 14.15 steps, it is 1.72× faster than the original model using 14 steps, while improving EditScore from 4.396 to 4.552.
Rebuttal Table SM-1: Component ablations and correction-only control. Mean steps, editing scores, and latency are measured on ImgEdit. The Speedup column is relative to original 64-step Lumina-DiMOO inference on ImgEdit (49.19 s/image). Background MSE is measured on PIE-Bench.