- Rating
- 2
- Confidence
- 4
- Informativeness
- 2
- Presentation
- 2
- Soundness
- 1
Read review
Paper summary
PhysFlow is a framework for physics-aware image editing that compresses textual reasoning and intermediate visual states into continuous latent representations. Built on Qwen-Image-Edit-2511, it uses a three-stage curriculum to learn explicit trajectories, progressively compress them, and adapt the image generator. The paper also introduces PhysTrace, a dataset of 11,448 annotated video trajectories. Experiments on PICABench and RISEBench report improvements over the backbone with modest additional inference cost, supported by automated evaluation and a user study.
Critical strengths
- The paper presents a coherent framework for incorporating physical transformation processes into image editing.
- The reported improvements are supported by complementary evaluation methods.
- The framework achieves a useful balance between editing quality and inference cost.
Critical weaknesses
- The incremental methodological contribution over existing latent reasoning approaches is not sufficiently established. Monet already explores visually informed latent reasoning and distillation, and is acknowledged in the paper. More directly, EndoCoT uses the same Qwen-Image-Edit-2511 backbone, recurrent hidden-state updates, and intermediate visual supervision, while avoiding intermediate image decoding at inference. Therefore, guiding image generation through continuous latent reasoning without rendering intermediate images does not, by itself, clearly distinguish PhysFlow.
- The experimental comparison does not adequately cover a closely related method for physics-aware image editing. PhysicEdit learns latent transition priors from videos for physical image editing without requiring intermediate video generation at inference. This directly relates to PhysFlow’s motivation concerning visual dynamics and computational efficiency. However, the paper primarily uses this work as the source of PhysicTran38K and does not provide a method-level comparison. In particular, the score of 53.82 in Tables 4 and 9 corresponds to PhysFlow trained on PhysicTran38K, rather than PhysicEdit itself. This dataset comparison therefore does not establish an advantage over the existing approach.
- The ablations do not sufficiently separate the benefits of dataset adaptation from those of structured latent reasoning. The no-reasoning configuration in Table 3 scores 54.26, matching the original backbone, and is described as the base model. The paper does not clearly report a direct fine-tuning baseline trained on the same PhysTrace source–target pairs with comparable generator adaptation and training budget. PICA100K-SFT cannot substitute for this control because it uses different data. Furthermore, the textual, visual, and joint reasoning ablations support the complementarity of the two modalities, but do not establish whether the progressive compression curriculum is more effective than learning latent conditioning with terminal supervision alone.
- The claim that general editing capabilities are preserved is stronger than the category-level results support. Although the overall ImgEdit-Bench score decreases only slightly, from 4.41 to 4.39, some categories show larger declines: Hybrid drops from 3.76 to 3.26, and Extract from 4.16 to 3.91. These results support broadly maintained aggregate performance, but not an unqualified claim of improvement without compromising general editing capabilities. This is a secondary limitation that calls for a more precise discussion of the trade-offs.
Questions for authors
- What is the central methodological contribution relative to EndoCoT and Monet?
- Was the no-reasoning configuration in Table 3 adapted using the same PhysTrace training data?
- How does performance change when the same recurrent latent tokens, final visual block, and trainable modules are retained, but training uses only terminal visual supervision and the generation loss?
- Could you provide a method-level comparison with PhysicEdit under a consistent evaluation protocol, explicitly accounting for backbone differences?
3 comments
Response to Reviewer xuAm
2. Comparison with PhysicEdit.
Thank you very much for pointing out the missing discussion of PhysicEdit.
As suggested, we have included a discussion of PhysicEdit in Section 2.3, “Latent Reasoning”, and Summary in Related Work.
Although both our proposed PhysFlow and PhysicEdit leverage textual explanations and visual trajectories to enhance the physical understanding of image editing models, they differ fundamentally in their underlying mechanisms, as detailed below.
(a) Our textual reasoning representation is different. PhysicEdit generates textual reasoning traces in the explicit semantic space of the MLLM, while separately modeling visual representations in the latent space. In contrast, our PhysFlow jointly performs textual and visual reasoning within a unified latent space. This formulation allows latent textual reasoning to guide visual-state prediction without generating explicit reasoning text at inference.
(b) Our visual reasoning explicitly models temporal dependencies. PhysicEdit jointly computes a set of transition-query representations, whereas PhysFlow autoregressively predicts successive visual-state representations conditioned on preceding latent textual explanations and visual states. This causal formulation explicitly captures the temporal dependencies underlying physical transitions.
(c) Learning unified latent reasoning requires a non-trivial training strategy. PhysicEdit trains its transition queries through feature alignment with projected DINOv2 and VAE features extracted from intermediate video frames. In contrast, our PhysFlow adopts a progressive training curriculum. It first supervises the prediction of all textual explanations and intermediate and final visual states using ground-truth text tokens and visual features extracted by the MLLM's visual encoder, respectively. It then progressively replaces the reasoning blocks for textual explanations and intermediate visual transitions with model-generated predictions, one at a time. Each replaced block receives indirect supervision through the original prediction loss applied to subsequent unreplaced positions, jointly with the diffusion loss. This stage encourages the model to learn causal dependencies across successive multimodal reasoning steps. Finally, the image generator is adapted to the learned latent representations using the diffusion loss.
We have added a quantitative comparison with PhysicEdit on PICABench-Superficial and ImgEdit-Bench in Table 1 and 5, respectively. Using the same Qwen-Image-Edit-2509 backbone, our PhysFlow achieves an overall score of 54.12% on PICABench-Superficial, outperforming PhysicEdit (52.35%) by 1.77 percentage points, with higher scores in seven out of eight categories. Moreover, our PhysFlow achieves an overall score of 4.31 on ImgEdit-Bench, compared with 4.05 for PhysicEdit, indicating stronger generalization to general image editing tasks.
| Method | LP | LSE | RFL | RFR | DFM | CSL | GST | LST | Overall |
|---|---|---|---|---|---|---|---|---|---|
| PhysicEdit (2509) | 51.62 | 61.63 | 51.56 | 52.96 | 44.03 | 45.89 | 57.64 | 53.07 | 52.35 |
| PhysFlow (2509) | 52.01 | 62.82 | 57.02 | 53.14 | 43.32 | 46.81 | 61.42 | 53.16 | 54.12 |
| Method | Add | Adjust | Extract | Replace | Remove | BackG | Style | Hybrid | Action | Overall |
|---|---|---|---|---|---|---|---|---|---|---|
| PhysicEdit (2509) | 4.31 | 4.17 | 2.34 | 4.45 | 4.44 | 4.03 | 4.74 | 3.36 | 4.85 | 4.05 |
| PhysFlow (2509) | 4.45 | 4.25 | 3.54 | 4.69 | 4.48 | 4.21 | 4.77 | 3.38 | 4.82 | 4.31 |
We have also added a comparison of inference latency between our PhysFlow and PhysicEdit in Table 4.
| Method | Latency (s) |
|---|---|
| PhysicEdit | 52.19 |
| PhysFlow (Ours) | 49.39 |
Response to Reviewer xuAm
3. Non-reasoning baseline trained on the same PhysTrace source–target image pairs.
Following your suggestion, we train a non-reasoning image editor on the same PhysTrace source–target pairs without reasoning annotations, using the same generator backbone (Qwen-Image-Edit-2511). Both methods use rank-512 generator LoRA and 537 generator updates, matching the generator adaptation configuration and budget. PhysFlow additionally trains its planner for 1,253 updates. We refer to this baseline as Qwen-Image-Edit-2511 + PhysTrace (w/o Reasoning) and report its results in Table 8.
As shown, the new baseline outperforms the original backbone by 0.86 percentage points. Under the same source-target image pair training setting, our curated PhysTrace achieves higher overall performance than PICA100K (i.e., Qwen-Image-Edit-2511 + PICA100K (w/o Reasoning)), highlighting its superior data quality and contribution to training efficiency. Furthermore, our latent reasoning framework further improves the overall score from 55.12% to 56.32% over this non-reasoning baseline, demonstrating the effectiveness of our proposed approach.
| Configuration | LP | LSE | RFL | RFR | DFM | CSL | GST | LST | Overall (%) |
|---|---|---|---|---|---|---|---|---|---|
| Qwen-Image-Edit-2511 | 53.29 | 62.98 | 56.78 | 57.66 | 44.16 | 45.99 | 59.37 | 55.34 | 54.26 |
| + PhysTrace (w/o Reasoning) | 55.07 | 65.16 | 57.96 | 56.67 | 42.71 | 45.30 | 63.23 | 53.81 | 55.12 |
| + PICA100K (w/o Reasoning) | 55.52 | 60.30 | 57.69 | 53.04 | 41.46 | 45.91 | 59.13 | 52.12 | 53.38 |
| PhysFlow | 54.45 | 65.19 | 58.52 | 57.99 | 44.91 | 47.31 | 64.32 | 56.43 | 56.32 |
4. Comparison between terminal-only supervision and full process supervision.
As suggested, we retain the same six recurrent latent tokens, 16-token final visual block, and trainable modules, and train the planner using only terminal visual supervision and the generation loss. With the same architecture and update budget, PhysFlow achieves 56.32% accuracy versus 55.57% for this baseline (Table 3(b)), a gain of 0.75 percentage points. These results support the benefits of textual reasoning supervision and dense intermediate visual supervision for improving physical consistency in image editing.
| Training strategy | LP | LSE | RFL | RFR | DFM | CSL | GST | LST | Overall |
|---|---|---|---|---|---|---|---|---|---|
| Terminal-only supervision | 53.84 | 64.48 | 57.73 | 59.18 | 42.57 | 46.68 | 63.59 | 55.71 | 55.57 |
| Full PhysFlow | 54.45 | 65.19 | 58.52 | 57.99 | 44.91 | 47.31 | 64.32 | 56.43 | 56.32 |
5. Additional analysis of general editing performance.
We value your suggestion to further investigate the category-level results more closely, particularly the declines in Hybrid and Extract despite the largely preserved overall performance on general image editing tasks.
We attribute these declines to the differences between physical-process editing and the specific requirements of these categories. Specifically, Hybrid involves combining multiple editing operations within a single instruction, while Extract requires isolating a subject while preserving its identity and visual details. In contrast, PhysTrace focuses on modeling causally connected state transitions for individual physical editing operations, guided by textual rationales and intermediate visual states (Sections 3.2–3.4). This mismatch in task focus may contribute to the observed performance degradation in Hybrid and Extract following physics-aware fine-tuning. We further observe that PhysicEdit and ChronoEdit, two other physics-aware editing methods, also exhibit relatively lower scores in Hybrid and Extract compared with other general editing categories in Table 5. We have included this analysis in Appendix C.1.
Response to Reviewer xuAm
We sincerely thank you for the time and effort you devoted to reviewing our manuscript and for your detailed, constructive feedback. We appreciate your recognition of our coherent framework, the complementary evaluations supporting our improvements, and our inference efficiency.
We have carefully considered each of your concerns and respond point by point below.
1. Central contribution relative to EndoCoT and Monet.
As suggested, we clarify the key contributions of our PhysFlow relative to EndoCoT and Monet.
1.1 Comparison with EndoCoT.
(a) Our latent representations receive richer process supervision. EndoCoT learns latent tokens for intermediate states through the diffusion loss, with the alignment loss applied only at the final state. In contrast, our PhysFlow progressively replaces explicit reasoning blocks with compact latent tokens, one block at a time. Specifically, these latent tokens receive indirect supervision from text prediction and visual alignment losses on subsequent uncompressed reasoning blocks, together with the diffusion loss computed on the final edited image. This provides richer supervision for learning compact and informative latent representations.
(b) Our diffusion supervision targets only the final edited image. During its initial training stage, EndoCoT learns latent tokens through diffusion losses computed on both the corresponding intermediate and final images, with the alignment loss applied at the final reasoning state. In contrast, PhysFlow supervises intermediate states in feature space and applies the diffusion loss only to the final edited image, reducing the diffusion computation required for intermediate-state supervision.
1.2 Comparison with Monet.
(a) Our textual and visual reasoning share a unified latent space. Monet solely distills visual evidence into latent embeddings while retaining explicit textual reasoning. In contrast, our PhysFlow jointly encodes textual physical explanations and visual trajectories within a unified multimodal latent space, without generating explicit reasoning text at inference.
(b) Internalizing multimodal reasoning requires a non-trivial training strategy. Monet adopts a three-stage supervised fine-tuning strategy: it first adapts the MLLM to interleaved textual-visual reasoning, then generates latent embedding targets in the second stage to supervise latent predictions in the third stage. In contrast, our PhysFlow employs a progressive compression mechanism without requiring a separate stage to generate latent embedding targets. Instead, the compressed latent tokens receive indirect supervision from text prediction and visual alignment losses on subsequent uncompressed reasoning blocks, together with the diffusion loss, to learn compact latent representations.
(c) Our framework targets reasoning-to-editing rather than reasoning-to-reasoning. Monet primarily focuses on enhancing the visual reasoning capabilities of MLLMs by reasoning over visual evidence directly within the latent visual space (reasoning-to-reasoning). In contrast, our PhysFlow aims to transfer physical knowledge from MLLMs into physics-aware image editing by explicitly modeling causal visual state transitions in the latent reasoning process (reasoning-to-editing).
We have included both EndoCoT and Monet in Section 2.3 of Related Work.