- Rating
- 2
- Confidence
- 3
- Informativeness
- 2
- Presentation
- 2
- Soundness
- 1(−2)
Read review
Paper summary
This paper studies how to allocate additional inference compute after an initial text-to-image generation fails. Motivated by the observation that resampling the original prompt is a strong baseline while prompt rewriting recovers a complementary subset of failures, the authors formulate post-failure recovery as a budget-allocation problem over stochastic exploration and semantic intervention. They propose FARE, which uses structured evidence from the initial failed image to distribute a fixed generation budget among original-prompt resampling and three failure-conditioned rewrite operators. FARE also employs a failure-referenced pruning mechanism that compares each candidate’s early denoising trajectory with the initial failed trajectory and terminates candidates likely to reproduce the same failure.
Critical strengths
-
The paper considers a relevant decision problem in post-failure text-to-image generation. Its experiments suggest that prompt rewriting is not always preferable to resampling and that the two interventions may recover partially different subsets of failed requests. These observations provide reasonable motivation for formulating recovery as a budget-allocation problem across intervention types.
-
The method is conceptually coherent. FARE separates the request-level decision of where to allocate computation from the trajectory-level decision of which candidates should continue receiving computation. The conservative fallback to all-resampling is well motivated given the strength of that baseline.
-
Several experimental choices are appropriate. In particular, images generated from rewritten prompts are still evaluated against the original prompt. The request-level train/validation/test split also avoids leakage among counterfactual allocations derived from the same request. The evaluation spans multiple benchmarks, generators, and inference budgets, and the appendix provides substantial implementation detail.
Critical weaknesses
-
Dependence on a single verifier. The same benchmark-specific verification pipeline is used to identify initial failures, derive failure features, construct allocation supervision, and determine whether recovery succeeds. Although this protocol demonstrates improvements under the selected verifier, it remains unclear whether the ranking of recovery policies is robust to alternative VLM evaluators or different success thresholds. Evaluation with an independent verifier and threshold-sensitivity analysis would help distinguish robust semantic improvements from evaluator-specific effects.
-
Limited coverage of generation quality. The evaluation focuses on semantic recovery rates and VQA scores, without a complementary measure of overall perceptual quality. Consequently, it is unclear whether improved satisfaction of individual semantic constraints is achieved at the expense of visual fidelity or aesthetic quality.
-
The claimed generality and compute efficiency are not yet established under a complete accounting of training and inference costs. The efficiency comparison primarily counts diffusion denoising steps. It does not provide unified end-to-end latency or cost accounting for failure diagnosis, LLM-based rewriting, semantic-preservation checking, trajectory storage, and trajectory-feature extraction. These omissions make it difficult to assess whether FARE offers a favorable practical trade-off relative to simple resampling.
Questions for authors
-
Does FARE retain its advantage when recovery is evaluated using an independent VLM backbone or an alternative benchmark-compatible verifier? In particular, does the relative ranking among FARE, resampling, and rewriting remain consistent across evaluators?
-
How sensitive are the reported recovery improvements to the semantic-success threshold? Results across a range of thresholds, or threshold-independent recovery curves, would help determine whether the gains are robust or concentrated near the selected decision boundary.
-
Does adaptive allocation affect visual fidelity or aesthetic quality? Could the authors report complementary reference-free quality metrics and compare the successfully recovered images produced by FARE with those produced by resampling and rewriting baselines?
-
What is the end-to-end latency and computational cost of FARE when failure diagnosis, prompt rewriting, semantic-preservation checking, trajectory storage, feature extraction, and diffusion generation are all included? How does this complete cost compare with simple resampling under the same hardware and batching conditions?
6 comments
Response to Reviewer CbcC (part 2/2)
Question 3
Does adaptive allocation affect visual fidelity or aesthetic quality? Could the authors report complementary reference-free quality metrics and compare the successfully recovered images produced by FARE with those produced by resampling and rewriting baselines?
Thank you very much for your constructive comment. We used three no-reference image quality assessment (NR-IQA) metrics—MUSIQ, LAION Aesthetic Score, and TOPIQ-NR—to directly evaluate the quality of the generated images. The results are reported below.
| Benchmark / Generator | N | MUSIQ ↑ | Aesthetic ↑ | TOPIQ-NR ↑ |
|---|---|---|---|---|
| GenEval2 / SD2.1 | 21 | 72.778 / 72.978 / 72.598 | 5.612 / 5.501 / 5.609 | 0.618 / 0.612 / 0.618 |
| GenEval2 / FLUX | 54 | 75.879 / 76.177 / 75.935 | 6.032 / 5.959 / 5.920 | 0.757 / 0.762 / 0.756 |
| TIFA / SD2.1 | 155 | 71.175 / 69.982 / 70.809 | 5.506 / 5.446 / 5.474 | 0.637 / 0.613 / 0.622 |
| CompBench / SD2.1 | 175 | 70.802 / 70.581 / 70.799 | 5.549 / 5.469 / 5.518 | 0.618 / 0.611 / 0.616 |
Note: Values are reported in the order Resampling / Train-selected Rewrite / FARE F2. N denotes the number of paired cases.
Across all four settings, FARE achieves image quality scores comparable to the baselines, suggesting that its adaptive recovery strategy does not substantially compromise visual quality.
Question 4
What is the end-to-end latency and computational cost of FARE when failure diagnosis, prompt rewriting, semantic-preservation checking, trajectory storage, feature extraction, and diffusion generation are all included? How does this complete cost compare with simple resampling under the same hardware and batching conditions?
We thank the reviewer for raising this important question. To provide a complete comparison, we measured the end-to-end runtime of FARE and Fixed Resampling on an NVIDIA A100 80GB GPU, with batch size = 1 and the same post-failure generation budget (B = 9).
Our measurements cover the complete recovery pipeline, including initial generation, failure diagnosis, prompt rewriting, semantic verification, trajectory storage, feature extraction and allocation prediction, recovery generation, and final recovery verification.
The results show that FARE without pruning takes 233.74 seconds per request, representing a 16.3% increase over Fixed Resampling (200.93 seconds). This additional computational cost brings recovery improvements of 1.1, 1.2, and 1.7 percentage points on GenEval2, TIFA, and T2I-CompBench, respectively. The average VQA score also increases from 0.2180 to 0.2243. With online pruning, FARE reduces the end-to-end runtime to 181.86 seconds, saving 9.5% compared with Fixed Resampling. The exploratory median checkpoint further reduces the runtime to 170.62 seconds, saving 15.1%. The detailed end-to-end latency and computational statistics are reported in Table 1, with a stage-wise breakdown provided in Table 2.
| Method | End-to-End Latency (s/request) | Latency Change vs. Resampling | Actual Denoising Iterations per Request¹ | Total Atomic VQA Calls² |
|---|---|---|---|---|
| Fixed Resampling | 200.93 | — | 500.0 | 470 |
| FARE without Pruning | 233.74 | +16.3% | 500.0 | 470 |
| FARE with Online Pruning (Retention-Constrained Checkpoint) | 181.86 | −9.5% | 386.0 | 321 |
| FARE with Online Pruning (Exploratory Median Checkpoint) | 170.62 | −15.1% | 379.7 | 310 |
| Stage | Resampling | FARE without Pruning | FARE with Pruning (Constraint Checkpoints) |
|---|---|---|---|
| Initial Generation | 11.86 | 10.71 | 11.98 |
| Failure Diagnosis | 7.00 | 7.40 | 7.93 |
| Prompt Rewriting | 0 | 20.08 | 20.65 |
| Semantic Verification | 0 | 14.67 | 14.97 |
| Allocation Feature Extraction and Prediction | 0 | 1.96 | 3.53 |
| Recovery Generation | 110.89 | 107.42 | 78.73 |
| Recovery Verification | 71.15 | 71.45 | 44.04 |
These results demonstrate that FARE without pruning trades additional computation for improved recovery performance, while online pruning can offset the additional processing overhead and achieve lower end-to-end computational cost than simple resampling.
Brief Summary of Responses to Reviewer CbcC
Dear Reviewer CbcC,
We sincerely thank you for your valuable time and constructive feedback. To facilitate a quick review, we summarize the key findings and additional experiments presented in our rebuttal below.
Weakness 1: Dependence on a Single Verifier
- Independent VLM Evaluation: We introduced two independent VLM evaluators, Molmo2-O-7B and InternVL3.5-8B, from different model families. The results provide additional evidence on evaluator sensitivity and show that relative recovery performance can vary across evaluators and generators. (See Weakness 1 for details.)
- Semantic-Threshold Sensitivity: We conducted additional experiments across five geometric-mean (GM) thresholds to examine the sensitivity of recovery performance to the semantic-success criterion. The results reveal generator-dependent trends, providing a more comprehensive assessment beyond the original threshold of 0.5. (See Weakness 1 for details.)
Weakness 2: Limited Coverage of Generation Quality
- Additional Visual Quality Evaluation: We introduced three reference-free image quality metrics, MUSIQ, Aesthetic Score, and TOPIQ-NR, to evaluate visual fidelity and aesthetic quality. The paired comparisons show generally small quality differences among FARE, resampling, and rewriting, without a consistent quality advantage for FARE. (See Weakness 2 for details.)
Weakness 3: Claimed Generality and Compute Efficiency Are Not Yet Established
- Cross-Generator Generalization: We conducted additional zero-shot transfer experiments to evaluate whether an allocator trained on SD2.1 can generalize to FLUX.1-dev without retraining. The results demonstrate transfer feasibility, although zero-shot FARE does not outperform resampling. (See Weakness 3 for details.)
- End-to-End Computational Cost: We conducted additional latency measurements on an NVIDIA A100 80GB GPU, using batch size 1 and the same post-failure generation budget. The measurements account for diagnosis, rewriting, semantic checks, feature extraction, diffusion generation, and verification, providing a more complete assessment of FARE's computational overhead relative to resampling. (See Weakness 3 for details.)
We hope these additional experiments and clarifications help address your concerns and provide a more comprehensive understanding of FARE's effectiveness and limitations. We would greatly appreciate your reconsideration of our submission and, if you find our responses satisfactory, an upward revision of your rating.
Thank you again for your valuable feedback and consideration.
Best regards,
The Authors
Response to Reviewer CbcC (part 1/2)
Question 1
We thank the reviewer for raising this important question. We agree that evaluating recovery with different VLM backbones is necessary to determine whether the reported improvements depend on a particular evaluator. To investigate this, we conducted additional evaluations using two independent VLMs from different model families: Molmo2-O-7B and InternVL3.5-8B. We kept the original test cases, recovery decisions, and generated images fixed. Neither evaluator participated in candidate generation, budget allocation, or image selection. Thus, only the evaluation backbone was changed. We compare Resampling, the train-selected rewriting baseline, and FARE F2 under both independent evaluators. The results are reported in the following.
The results lead to two observations. First, FARE F2 maintains competitive recovery performance across both independent evaluators, suggesting that its effectiveness is not solely attributable to the original benchmark evaluators. In particular, FARE F2 achieves the highest recovery rate in most of the evaluated benchmark and evaluator combinations. Second, the relative ranking of the methods is not perfectly consistent across evaluators. In some settings, Resampling or the fixed rewriting baseline performs better than FARE F2. These results suggest that evaluator-dependent variations exist, which is expected given the differences in semantic judgment across VLM backbones. Nevertheless, FARE remains consistently competitive and achieves the best performance in most settings.
Question 2
Thank you for this suggestion. We evaluated threshold sensitivity on the same held-out GenEval2 cases—152 for SD2.1 and 131 for FLUX—while keeping allocation decisions, generation budgets, and final selected images fixed. We swept the geometric-mean semantic-success threshold over {0.3, 0.4, 0.5, 0.6, 0.7}, without retraining, regenerating images, or selecting a threshold based on test performance. We report both recovery-rate curves and FARE-minus-baseline differences.
Under the original evaluator, FARE F2 outperformed resampling on FLUX at all five thresholds, with gains of 3.05, 1.53, 3.82, 5.34, and 6.11 percentage points, respectively. Thus, this advantage was not confined to the selected boundary of 0.5. On SD2.1, however, the differences ranged from −0.66 to +0.66 percentage points, indicating a small and threshold-sensitive advantage. FARE also did not consistently outperform the stronger Train-selected Best baseline on FLUX.
Additionally, the evaluation with Molmo2-O-7B and InternVL3.5-8B showed positive FARE-versus-resampling point estimates on FLUX across all five thresholds, but no consistent advantage on SD2.1.