- Rating
- 2
- Confidence
- 3
- Informativeness
- 2
- Presentation
- 3
- Soundness
- 2
Read review
Paper summary
The authors observe that existing parameter-efficient continual learning methods focus on reducing the number of trainable parameters while overlooking the impact of training iterations and optimization strategies on training efficiency. To address this limitation, they propose an adaptive training budget allocation framework for parameter-efficient continual learning. The proposed method significantly reduces the number of training iterations with only a minor loss in accuracy.
Critical strengths
- The authors investigate parameter-efficient continual learning from the perspectives of reducing training iterations and update allocation. This novel perspective offers significant research value and has far-reaching implications for future research.
- The experimental results are impressive, demonstrating that the proposed method reduces the number of training iterations required by existing parameter-efficient continual learning methods.
Critical weaknesses
- I believe the proposed ‘HowMuch’ component has strong practical value. However, for the ‘Where’ component, the authors design three different allocation strategies for three distinct CL methods, indicating that ‘Where’ relies heavily on the specific architecture and optimization structure of each method. This raises concerns about whether ‘Where’ can truly serve as a generalizable training budget allocation framework.
- A reduction in the number of training iterations does not necessarily translate into a reduction in overall training time. The proposed framework introduces additional computational overhead from probe statistics, loss-gradient routing, class sampling, or classifier-head selection, which may increase the time required per epoch.
Questions for authors
- As mentioned in Weakness 1, the ‘Where’ component relies on the specific architecture and optimization structure of each continual learning method to allocate update opportunities. For example, EWC-LoRA (ICLR 2026), a regularization-based continual learning method, does not have the dual-branch structure of CL-LoRA or the classifier-alignment interfaces used in MOS/TUNA. Would AllocAdapt remain effective when applied to such methods, and how much method-specific modification would be required? Does this imply that a new Where strategy must be designed whenever AllocAdapt is extended to a new continual learning method?
- Regarding Weakness 2, could the authors report the overall training time of AllocAdapt and the original methods under the same hardware conditions, including the additional computational overhead introduced by g probe statistics, loss-gradient routing, class sampling, or classifier-head selection? This would help determine whether the substantial reduction in training iterations translates into meaningful training speedups in practice.
- Equation (2) assumes that a weighted combination of R, G, and I can represent training demands. How can the authors demonstrate that the demand score defined in Equation (2) accurately reflects the actual optimization needs?
4 comments
Shared structure of Where and evidence for adaptive allocation
Thank you for your questions about the generality of Where. We would like to make explicit which elements are shared and how they connect to the different training interfaces.
1. Shared allocation structure
Where follows a common decision structure: identify allocation units in the native training process, obtain optimization feedback, and assign training opportunities under the horizon selected by HowMuch and the constraints of that interface.
| Learner | Allocation decision | Training feedback | Execution constraint |
|---|---|---|---|
| CL-LoRA | When to switch from joint training to separated CE/KD routes | CE/KD improvement and KD rebound | Switching occurs within the assigned horizon and admissible decision interval |
| MOS | How to distribute synthetic features across classes | Class-gradient statistics | Fixed total feature budget and balanced target objective |
| TUNA | How frequently to update classifier parameter blocks | Block-gradient statistics | A fixed number of active blocks per routed step, with coverage scheduling |
The common contribution is this organization of training allocation: feedback informs the distribution of work within an assigned horizon. The learner-specific interface determines how that decision is executed. All three integrations retain the base architecture and inference procedure.
For another learner, the integration requires specifying its allocation units, observable training signals, and admissible actions. For a regularization-based learner, existing parameter groups and gradients of the native regularized objective offer a possible interface. Its effectiveness would need to be evaluated for that integration.
2. Interpretation and empirical role of the demand score
Regarding the demand score, captures residual loss relative to the probe's starting loss, captures continued improvement, and captures recent loss increases. These signals distinguish training trajectories with similar endpoint losses but different ongoing progress or instability. Historical normalization interprets them relative to earlier tasks of the same stage.
The component comparisons provide evidence about the resulting allocation decisions. Full has the highest reported Avg and Final in all four groups in Table 2 at iteration budgets comparable to the component variants. For MOS specifically, adding HowMuch to Where-only improves Final on both evaluated datasets while using fewer iterations. The recorded trajectories additionally connect calibrated demand to executed horizons.
3. Practical efficiency including allocation work
Practical efficiency gains are observed across all nine learner–dataset settings, with measured training-time reductions of 13.97–54.80%, including probing and allocation work.
The supplementary revision summary identifies the corresponding additions. We hope these clarifications make the shared allocation structure and its empirical support easier to assess.
Thank you for the author's reply. The concerns about generalizability are not fully resolved. For regularization-based learners, the author only proposed one possible integration method without actual verification. For Eq.(2), I understand the design motivation, but the rationality and reliability of the score still lack direct verification. Therefore, I'm keeping my socre.
Thank you for your follow-up. We would like to clarify the motivation for selecting the three base learners and the experimental evidence supporting demand-guided allocation.
1. Competitive base learners and the scope of generalization
AllocAdapt is an allocation framework built on existing continual learners. Its contribution is to determine how much training to perform and where to allocate updates, while preserving the recognition capabilities of the underlying learners. Implementing the framework on established methods therefore directly serves our research objective.
We selected CL-LoRA, MOS, and TUNA because they combine competitive recognition performance with different optimization interfaces. Their published benchmark results illustrate the relevance of this choice. For example, the following results are reported in Table 1 of the TUNA paper. Each entry denotes Avg / Final accuracy (%).
| Method | CIFAR-100 | ImageNet-R | ImageNet-A |
|---|---|---|---|
| EASE | 91.51 / 85.80 | 81.74 / 76.17 | 65.34 / 55.04 |
| SLCA | 92.49 / 88.55 | 81.17 / 77.00 | 68.66 / 58.74 |
| SSIAT | 93.52 / 90.07 | 83.20 / 78.85 | 70.83 / 62.23 |
| RanPAC | 94.00 / 90.62 | 82.98 / 77.94 | 69.32 / 61.82 |
| MOS | 93.30 / 89.25 | 82.96 / 77.93 | 67.08 / 56.22 |
| TUNA | 94.44 / 90.74 | 84.22 / 79.42 | 73.78 / 64.78 |
Source: TUNA, Table 1. These experiments use 20 tasks for CIFAR-100 and 10 tasks for ImageNet-R and ImageNet-A.
The comparison with the regularization-based learner mentioned in your review is also informative. Table 2 of the EWC-LoRA paper reports the following results under its own 10-task evaluation protocols:
| Method | CIFAR-100 | ImageNet-R | ImageNet-A |
|---|---|---|---|
| CL-LoRA | 92.25 / 87.65 | 85.20 / 78.72 | 70.76 / 57.62 |
| EWC-LoRA | 92.27 / 87.91 | 78.95 / 72.86 | 68.33 / 59.89 |
Source: EWC-LoRA, Table 2. Comparisons should be made within each table because the source papers use different experimental protocols.
In particular, CL-LoRA exceeds EWC-LoRA by 6.25 percentage points in Avg and 5.86 points in Final on ImageNet-R, and by 2.43 points in Avg on ImageNet-A. These results demonstrate the competitiveness of the selected learners relative to a recent regularization-based approach on these benchmarks, motivating the study of their training efficiency.
Beyond baseline accuracy, our three implemented integrations cover loss-gradient routing in CL-LoRA, class-sample allocation in MOS, and classifier-block updates in TUNA. All three are evaluated on three datasets. The shared contribution is the probe-and-calibrate allocation framework, instantiated through each learner’s available update interface. This provides experimental evidence across several distinct interfaces within the evaluated setting.
2. Evidence for demand-guided horizon selection
Equation (2) combines complementary training signals: residual loss, late-probe progress, and instability. These signals are calibrated against earlier tasks within the same training stage to guide horizon selection.
The component experiments provide evidence across all four learner–dataset comparisons in Table 2 of our manuscript. Full achieves higher Avg and Final accuracy than Where-only while using the same or fewer iterations. Both variants enable Where, so this comparison evaluates the additional contribution of adaptive horizon selection.
| Learner / Dataset | Where-only Avg / Final | Full Avg / Final | Final improvement |
|---|---|---|---|
| CL-LoRA / CIFAR-100 | 89.679 / 83.47 | 90.160 / 84.56 | +1.09 |
| CL-LoRA / ImageNet-R | 81.556 / 73.45 | 81.821 / 74.63 | +1.18 |
| MOS / CIFAR-100 | 92.826 / 88.66 | 92.852 / 88.88 | +0.22 |
| MOS / ImageNet-R | 84.110 / 78.60 | 84.129 / 78.63 | +0.03 |
Accuracy is reported in percent; improvements are in percentage points.
For example, MOS uses the score in Eq. (2). Adding HowMuch improves both accuracy metrics on CIFAR-100 and ImageNet-R while reducing iterations. The corresponding CL-LoRA comparisons also show improvements, using its learner-specific demand mapping.
Together, these results provide component-level evidence for the practical utility of demand-guided horizon selection. MOS offers an example directly relevant to the controller using Eq. (2), while CL-LoRA provides complementary evidence for the broader probe-and-calibrate allocation principle.
We have revised the manuscript to include training-time measurements, multi-seed statistics, and additional implementation details. A summary of the manuscript revisions is available in the supplementary material.
A Gentle Request for Feedback on Our Rebuttal
Dear Reviewer,
Thank you again for your constructive feedback. We have provided point-by-point responses to all the questions and concerns raised in your review and conducted additional experiments. We have also revised the manuscript to incorporate these results and clarifications and uploaded the updated PDF to OpenReview.
We would appreciate it if you could review our responses and the revised manuscript and let us know whether any concerns remain. We would be happy to provide further clarification and would welcome your updated assessment in light of these additions.
Thank you for your time and consideration.
Best regards, The Authors