- Rating
- 2
- Confidence
- 3
- Informativeness
- 2
- Presentation
- 2
- Soundness
- 2
Read review
Paper summary
This paper observes category-dependent visual modes in pretrained feature spaces and proposes adaptive proxy modeling to estimate their number via graph Laplacian spectra, constructing adaptive proxies without additional trainable parameters.
Critical strengths
The paper is well motivated, with a novel perspective on category-dependent intra-class structures and a clear, logical narrative.
Critical weaknesses
- The experimental comparisons are not entirely fair, particularly for ReID, where the reported gains may largely stem from the stronger backbone. The ablation results also suggest that APM provides limited improvements, while the baseline already outperforms most competing methods.
- The experiments lack comparisons with related methods that also use DINOv2 features and proxy-based learning, as discussed in the related work.
- Table 4 lacks sufficient baselines. Although the ImageNet-1K results appear promising, the only comparison is with naive full fine-tuning. Methods such as Proxy-FDA, specifically designed to preserve linear-probing performance, are missing, weakening the claim of better representation preservation.
Questions for authors
My main concern is the fairness of the comparisons, as many experimental comparisons are not sufficiently meaningful or clearly justified. I would like to see evaluations on more datasets and more comprehensive, clearly structured ablation studies.
3 comments
Response to Reviewer 4x5k, W2
Response to W2:
Comparison with Methods in Related Work. The table below summarizes which DINOv2-based FGVR methods and proxy-based methods discussed in Related Work we compare with and explains the exclusions.
| Type | Method | Compared? | Reason |
|---|---|---|---|
| DINOv2-based FGVR method | PDiscoFormer (Aniraj et al., 2024) | yes | - |
| DINOv2-based FGVR method | IVPT (Wang et al., 2026b) | yes | - |
| DINOv2-based FGVR method | Prompt-CAM (Chowdhury et al., 2025) | no | Uses a frozen DINOv2 backbone for interpretable trait localization, which differs from our full fine-tuning setting |
| proxy-based method | Proxy-FDA (Huang et al., 2025) | no | Uses a different evaluation protocol, and no official public implementation is available for reproducing it under our setting |
| proxy-based method | Other methods cited in Related Work | no | Their published evaluations use different backbones or task settings |
Response to Reviewer 4x5k, W3
Response to W3:
Comparison with Existing Preservation Methods on CUB-200-2011. We agree that Table 4 should include stronger preservation-oriented baselines. We therefore add L2-SP and LP-FT, two methods designed to better retain pretrained representations, using DINOv2-B/14 under the same CUB-200-2011 fine-tuning and ImageNet-1K linear-probing protocol.
| Configuration | CUB Acc. (%) | ImageNet-1K Linear Probe Acc. (%) | Total Gain (pp) |
|---|---|---|---|
| Pretrained | - | 83.6 | - |
| Full FT | 89.2 (+0.0) | 79.7 (+0.0) | 0.0 |
| L2-SP | 90.1 (+0.9) | 82.4 (+2.7) | +3.6 |
| LP-FT | 90.2 (+1.0) | 82.6 (+2.9) | +3.9 |
| Ours | 91.4 (+2.2) | 81.5 (+1.8) | +4.0 |
L2-SP and LP-FT retain more ImageNet-1K linear-probe accuracy, while APM achieves the highest CUB accuracy. Compared with full fine-tuning, APM improves CUB by +2.2 points and ImageNet linear probing by +1.8 points, yielding a total gain of +4.0 points. This indicates that APM does not simply maximize preservation of the pretrained representation; rather, it uses category-specific pretrained structure to improve downstream adaptation while retaining more pretrained knowledge than standard full fine-tuning.
Moreover, thank you for suggesting Proxy-FDA. Proxy-FDA is relevant, but no official public implementation is available for reproduction under the same setting for a fair comparison. We therefore use L2-SP and LP-FT as controlled preservation baselines, while Proxy-FDA is already cited in our manuscript.
Response to Reviewer 4x5k, W1
Response to W1:
We thank the reviewer for raising this concern. We agree that comparisons across methods with different backbone architectures should be interpreted cautiously. Our main evidence for the contribution of APM therefore comes from controlled comparisons using the same DINOv2 backbone and training setup, as discussed below:
Baseline Strength and Fairness of Comparison. The 2-cls baseline is indeed strong and competitive with several existing ReID methods, whereas the 1-cls baseline remains below many of them. Importantly, APM is evaluated on top of both settings. Under the 1-cls setting, APM improves MSMT17 from 70.9/86.9 to 79.5/91.5 in mAP/Rank-1 and Market-1501 from 85.0/93.0 to 93.3/96.4. Since the backbone and training configuration remain unchanged, these gains isolate the contribution of APM rather than the effect of using a stronger backbone.
APM also improves the stronger 2-cls baseline. On MSMT17, performance increases from 73.3/89.0 to 80.3/91.6, corresponding to gains of +7.0 mAP and +2.6 Rank-1. On Market-1501, where the baseline already reaches 92.0/96.0, APM further improves performance to 93.5/96.9. Thus, APM provides clear gains both when the baseline is relatively weak and when it is already competitive.
These controlled comparisons also complement our ablation study, where adaptive proxy modeling outperforms both single-proxy and fixed multi-proxy alternatives under the same backbone and optimization setting. This further supports that the gains come from the proposed adaptive modeling rather than from backbone strength alone.
Further Comparison with DRFormer. To provide a broader comparison with methods using different architectures, we additionally include DRFormer in Table 2 of the revised manuscript. DRFormer combines CLIP and DINOv2 encoders with multiple learnable tokens, while our model uses only a single DINOv2 encoder. Despite this simpler backbone configuration, our 2-cls model achieves higher mAP on both Market-1501 (93.5% vs. 92.9%) and MSMT17 (80.3% vs. 78.7%).