- Rating
- 2
- Confidence
- 2
- Informativeness
- 2
- Presentation
- 2
- Soundness
- 2
Read review
Paper summary
The paper proposes Generative Inference Clock Optimization (GICO), an offline method for learning sampling-time allocations for frozen generative ODEs. It represents schedules as normalized densities, uses measured terminal utilities to train a surrogate, and learns an anchored clock policy conditioned on context, solver, and NFE. Experiments cover temporal, molecular, and image generation, including held-out-budget evaluations.
Critical strengths
The density representation separates allocation from node count, enabling a shared formulation across inference budgets. Utility-based supervision targets downstream quality directly, with theoretical analysis explaining how local errors affect terminal utility. Experiments span multiple domains and include representation, context, and reference-selection controls.
Critical weaknesses
- The surrogate determines reference weights, guides refinement, and supports checkpoint selection. Its calibration assumptions are not verified at the anchor and refined allocations. Accuracy on held-out reference clocks does not establish reliability at policy-generated clocks, where optimization may exploit prediction errors.
- For temporal and molecular tasks, each context is primarily assigned one reference clock plus a uniform anchor; ImageNet similarly assigns each class to one density. The surrogate subsequently scores the full reference pool for each setting. The paper does not sufficiently validate these unobserved context-clock predictions using a directly measured multi-clock evaluation panel.
- The pool is explicitly specified but remains manually configured. Its composition, exponents, reversals, and cross-model schedule mappings may affect both surrogate coverage and policy anchors.
- In the full text-to-image comparison, transfer combines 12,800 observations from source NFEs, whereas direct fitting uses 6,400 observations. Excluding the target NFE establishes a valid transfer setting, but superiority over direct fitting cannot be attributed solely to budget generalization.
- Its unbiased estimator provides a reasonable motivation, and evaluation with other metrics is useful. However, unbiasedness does not establish low variance or reliable clock rankings. The paper does not adequately assess supervision noise, ranking stability, or sensitivity to the choice of training metric.
- The distinction between distributions over whole clocks and densities allocating evaluations over time could also be presented more clearly.
Questions for authors
- Can you measure surrogate calibration and ranking accuracy at the weighted anchors and policy-generated clocks? How do these errors relate to realized utility gains, and how sensitive are the results to the tilt temperature and refinement coefficient?
- Please evaluate multiple reference clocks on the same held-out contexts or classes.
- What motivated the reference-pool composition and parameter choices?
- Please compare transfer and direct fitting using equal total observation budgets.
- Why was KID selected as the primary image supervision metric?
- Can you report repeated-estimation uncertainty, clock-ranking stability, and agreement with final FID rankings, or compare alternative supervision metrics under matched collection costs?
please engage
Dear reviewers,
Please join the discussion with the authors.
Best, AC