- Rating
- 2
- Confidence
- 2
- Informativeness
- 2
- Presentation
- 2
- Soundness
- 2
Read review
Paper summary
This paper proposes a good question. Calibrating each module independently keeps intermediate states physically meaningful but leaves large terminal errors, whereas end-to-end training reduces terminal errors by letting moduels compensate for one another's mismatch. Authors treat simulator as differentiable computation graph and embeds a learnable correction at an internal node.
Critical strengths
- This is really a good question.
- Error propagation, gradient alignment and palcement analysis have clear meaning.
Critical weaknesses
- The error is only a Lipchitz expansion. Is there any mechanism that regulaize each module's dynamic?
- Under my understanding, this paper only adds a leanable small network as regulaization. This small network absorb mistakes. The mechanism seems too simple. The regualization only comes from errors, not from real dynamics.
Questions for authors
See Weaknesses.
3 comments
3. Experimental comparison with existing correction and regularization methods.
We added three representative baselines: APHYNITY (loss regularization), NeuralSim (simulator augmentation), and Solver-in-the-Loop (state correction), adapted to CTM under matched data splits, inputs, terminal supervision, and training budgets.
Results at 50% injected bias, with correction at where applicable, are summarized below. Mean RMSE equally weights terminal and internal RMSE.
| Method | Terminal RMSE | Internal RMSE | Mean RMSE |
|---|---|---|---|
| APHYNITY | 392.20 | 429.21 | 410.71 |
| NeuralSim | 477.05 | 448.57 | 462.81 |
| Solver-in-the-Loop | 277.60 | 910.67 | 594.14 |
| GraphBC | 286.54 | 449.80 | 368.17 |
GraphBC achieves the best overall balance: terminal accuracy close to Solver-in-the-Loop (286.54 vs. 277.60), but with approximately 51% lower internal RMSE. It also achieves the lowest Mean RMSE, outperforming APHYNITY by 10.4%.
The expanded Table 2 confirms the lowest Mean RMSE across all six bias–placement settings. Implementation details and additional results are provided in the appendix.
4. Dynamics-embedded and dynamics-coupled correction in GraphBC
We clarify that GraphBC does not impose an explicit dynamics regularizer on every module. Instead, it embeds learnable corrections within the original simulator and trains them through downstream dynamical propagation. The mechanism consists of three aspects:
(a) Dynamics-embedded correction. GraphBC retains the original dynamical modules and introduces learnable residual corrections to compensate for mismatches beyond the original model's representational capacity:
The correction complements rather than replaces the original dynamics, extending the model's expressiveness while retaining its underlying computational structure.
(b) Dynamics-coupled training. Correction networks are trained using the terminal loss, with gradients backpropagated through the original downstream dynamical modules:
(c) Graph-based characterization of correction–dynamics interactions. GraphBC uses module dependencies and sensitivities to distinguish local model mismatch from downstream propagated errors. Its gradient-alignment analysis further establishes conditions under which terminal-loss updates also improve intermediate-state fidelity. Together, these analyses explicitly characterize the interaction between learned corrections and the original dynamical system.
The supervision comes from prediction errors, but the correction updates are shaped by the downstream dynamics, rather than independent local error fitting.
Thanks for your reply. I acknowledge the comprehensive ablation study can address some of my concern. But I really cannot support a higher score for the novelty in ICLR, such a top-tier conference. I prefer to maintain my original score.
Response to Weakness 1 and Weakness 2
Dear Reviewer QWaM:
Thank you for your thoughtful comments. We acknowledge that our methodology was not clearly presented and further clarify it through a detailed comparison with dynamics regularization below.
2. GraphBC introduces explicit model corrections rather than relying solely on dynamics regularization.
We thank the reviewer for raising this important point. Unlike dynamics regularization, which constrains model learning through additional loss terms, GraphBC introduces learnable corrections directly into the computational graph, extending the model's representational capacity . We further clarify the advantages of GraphBC in the following comparison:
The three key advantages of GraphBC are: