- Rating
- 3
- Confidence
- 4
- Informativeness
- 2
- Presentation
- 3
- Soundness
- 2
Read review
Paper summary
FlashAR converts pretrained raster-order image generators into diagonal-parallel generators through post-training. It retains the horizontal prediction pathway, adds a vertical branch from a pre-final layer, and learns a gate to combine the two predictions. Adaptation first trains the added components and then finetunes the full model. Experiments cover LlamaGen, Emu3.5-Image, and the autoregressive component of GLM-Image.
Critical strengths
-
The paper demonstrates a substantial speedup against an optimized raster baseline.
-
Reusing pretrained models makes the approach practically useful. Across four LlamaGen scales, 25 epochs of adaptation improve FID over the original checkpoints and outperform the reported BlockDiffusion results. The branching, gating, and training-schedule ablations provide evidence that the adaptation choices each help.
Critical weaknesses
-
Diagonal two-direction decoding is inherited from NAR, so the main contribution is how it is adapted to pretrained models. The "naive adaptation" baseline in Figure 4 is not specified precisely enough to establish the gain over a straightforward NAR-style finetuning baseline. The ablations are also confined to LlamaGen-L, leaving the importance of these components in the main 34B experiment untested.
-
The adaptation cost claim needs better accounting. The paper reports approximately 80K image-text pairs, 50K optimizer steps, and 80M post-training tokens, but does not state the text-to-image global batch size or distinguish unique dataset tokens from total token presentations. Without that distinction, the claim of using a small percent of pretraining tokens is difficult to interpret as a training-cost comparison.
Questions for authors
-
What exactly changes in the naive-adaptation baseline, including branch placement, attention mask, fusion rule, and training schedule? Is it equivalent to finetuning a NAR-style model from the same raster checkpoint?
-
Does the 80M-token figure count unique corpus tokens or all tokens processed during optimization? Please report the global batch size, total token presentations, and GPU-hours for both adaptation stages.
2 comments
Response to Rebuttal
Thanks for the answers and new experiments. The baseline details and training costs clear up most of my questions. Please make it clear that 0.05% counts unique tokens, not all tokens used during training. The 34B tests help, but gating and the two-stage training are still not tested separately there. I will keep my score at 3.
We thank the reviewer for the positive assessment, supportive rating, and constructive questions regarding the Naive Adaptation baseline specification, 34B component ablations, and token/cost accounting.
Q1: What exactly changes in the naive-adaptation baseline, including branch placement, attention mask, fusion rule, and training schedule? Is it equivalent to finetuning a NAR-style model from the same raster checkpoint? And how do these components perform on the 34B model?
A1: (1) Exact specification of the Naive Adaptation baseline (Fig. 4b): Yes, the Naive Adaptation baseline in Fig. 4(b) is equivalent to fine-tuning a standard NAR-style dual-head model initialized from the pre-trained raster-scan checkpoint. In Fig. 4(b) (evaluated on LlamaGen-L at fixed CFG=1.5 after 25 epochs), each variant modifies one component at a time:
At CFG=2.0, full FlashAR reaches FID 3.16. We will add this explicit specification to Section 4.3.
(2) Component ablations on Emu3.5-Image-34B (L=64): To verify the importance of our design choices on the 34B T2I backbone, we ablate the branch point m and the auxiliary directional loss weight λaux on Emu3.5-Image using PickScore [1]:
m=48m=56m=60(Pre-final, Default)m=64(Final Layer, NAR-style)Consistent with LlamaGen-L, attaching the vertical head directly to the final layer (m=64) or removing the auxiliary directional supervision (λaux=0) substantially degrades performance on Emu3.5-Image. We will include these 34B ablations in the revised manuscript.
Q2: Does the 80M-token figure count unique corpus tokens or all tokens processed during optimization? Please report the global batch size, total token presentations, and GPU-hours for both adaptation stages.
A2: Thank you for pointing out this important distinction. The 80M tokens figure reported in the paper counts unique corpus visual tokens (∼80,000 unique image-text pairs×1,024 visual tokens/image≈81.92M unique visual tokens at 512×512 resolution, corresponding to 0.05% of Emu3.5's 150B pre-training tokens). Below we report the complete accounting across both adaptation stages, including the global batch size, total token presentations, peak GPU memory, and GPU-hours:
Even when measured by total visual token presentations (409.6M tokens across 50,000 steps at Global Batch Size = 8), our post-training budget represents only 0.27% of Emu3.5's 150B pre-training token budget (and 0.05% in unique corpus tokens). We will explicitly report both unique dataset tokens and total token presentations alongside the global batch size and GPU-hours in the revised manuscript.
Reference:
[1] Pick-a-Pic: An Open Dataset of User Preferences for Text-to-Image Generation