- Rating
- 4(+1)
- Confidence
- 4
- Informativeness
- 3
- Presentation
- 3
- Soundness
- 2
Read review
Paper summary
This paper proposes a decision boundary aware framework for open set graph domain adaptation. It employs virtual class learning to enhance source domain discriminability, followed by mining class relation to distinguish known and unknown target nodes. Moreover, it employs contrastive learning with perturbations to facilitate cross domain knowledge transfer. Both theoretical analyses and experiments demonstrate the effectiveness of the framework.
Critical strengths
- The paper addresses the practically relevant OSGDA problem, where the target domain contains novel classes absent from the source domain.
- The method improves knowledge transfer by optimizing source domain decision boundaries, rather than focusing solely on unknown class separation in the target domain.
- Virtual class learning is well suited to OSGDA, as it enhances known class discriminability and supports the separation of known and unknown classes.
- The paper provides theoretical analyses to support the proposed virtual class learning mechanism, offering theoretical insights.
Critical weaknesses
- This paper adopts overall accuracy and HS as the evaluation metrics for OSGDA, which focuses on the overall performance on known and unknown classes. It is unclear how the proposed method performs on known class classification and unknown class detection, respectively, in OS* (known accuracy) and AUC, compared with OSGDA baselines.
- The unknown class identification mechanism relies on the assumption that unknown target nodes generally exhibit larger KL divergence from source domain class prototypes. However, this assumption may not always hold under substantial domain shifts or when unknown classes are semantically similar to known classes. In such cases, the divergence distributions of known and unknown nodes may overlap significantly.
- For threshold selection, Eq. (12) employs a fixed threshold of to select high confidence known nodes. However, different datasets may have different representation distributions. The paper does not justify this threshold or investigate its sensitivity, leaving its general applicability unclear. In addition, in Eq. (9), it is unclear how to “adaptively determine the threshold according to the underlying data distribution”.
Questions for authors
- How the proposed method performs on known class classification and unknown class detection, respectively, compared with the OSGDA baselines? Please provide comparison between the proposed method and OSGDA baselines in OS* (known accuracy) and AUC on some representative tasks.
- How do improved source domain decision boundaries facilitate the recognition of unseen target classes and under what conditions is this mechanism effective?
- This paper uses a fixed -quantile of a fitted beta distribution to identify the unknown nodes, how does the fixed quantile parameter affect unknown-node identification when the proportion of unknown nodes varies across target domains?
- In Eq. (9), how to adaptively determine the threshold according to the underlying data distribution?
- Why do the authors adopt a single beta distribution to model the KL divergence values of both known and unknown target nodes? Can a single beta distribution adequately model the divergence values of both known and unknown nodes, compared with alternatives such as gaussian mixture models?
- For the edge down-weighting strategy, if known nodes are incorrectly identified as unknown, their associated edges will also be down-weighted, potentially suppressing useful neighborhood information. How does this affect subsequent message passing, and can the proposed method mitigate this issue?
5 comments
Q3: Discussion About the Impact of the Fixed -quantile to the Unknown-class Nodes.
R3: Thank you for your insightful question! Although the quantile parameter is fixed, the threshold is dynamically determined by the fitted Beta distribution of KL divergence values in each target domain, allowing it to adapt to different divergence distributions. controls the strictness of unknown-class nodes identification. A smaller lowers the threshold and identifies more nodes as unknown, potentially improving unknown-class recall while increasing false positives. Conversely, a larger imposes a stricter threshold, reducing false positives while potentially missing some unknown nodes. Thus, when the proportion of unknown nodes varies across target domains, we set a moderate to control the trade-off between unknown-class precision and recall. Furthermore, in Section 3.4 we conduct sensitivity analyses with in , as shown in Figure 4. The results indicate that achieves a favorable balance in overall ACC and HS across the evaluated transfer tasks.
We hope this explanation adequately addresses your concerns.
Q4: Discussion About How to Adaptively Determine the Threshold.
R4: Thank you for your valuable question! We adopt a distribution-aware modeling strategy to derive an adaptive identification threshold rather than relying on a manually specified divergence threshold. The key insight is to decouple the selection criterion from the actual divergence threshold. The fixed quantile parameter specifies the proportion of nodes selected from the upper tail of the distribution, while the fitted Beta distribution determines the corresponding threshold based on the underlying divergence distribution. Consequently, the threshold adapts to variations in target-domain data distributions while maintaining a consistent selection criterion.
Furthermore, we have clarified the rationale behind this distribution-aware thresholding strategy in Section 2.4 of the revised manuscript. We hope this explanation adequately addresses your concern.
Q5: Discussion about the Beta Distribution.
R5: Thank you for the comment! We conduct experiments to explore the effects of both GMMs and the Beta distribution for modeling the calculated KL divergence values. The experimental results (ACC/HS) are presented below:
| Tasks | Beta Distribution Model | Gaussian Mixture Model |
|---|---|---|
| A2C | 75.33 1.19 / 74.43 1.35 | 70.22 0.99 / 64.17 0.99 |
| A2D | 73.83 1.70 / 71.33 2.26 | 69.08 1.89 / 60.75 2.96 |
| C2D | 74.23 2.92 / 70.60 5.52 | 69.49 2.46 / 59.63 4.66 |
| C2A | 66.89 4.63 / 64.72 5.90 | 60.80 3.97 / 53.17 4.70 |
| D2A | 59.21 1.77 / 53.58 4.71 | 54.33 1.73 / 44.58 4.59 |
| D2C | 68.64 1.19 / 65.07 2.81 | 61.98 0.92 / 53.91 1.60 |
Our experiments show that Beta distribution modeling consistently outperforms GMM-based modeling across all six transfer tasks, especially in HS. Through further analysis, we identify the underlying reasons for this discrepancy. First, the calculated KL divergence values are intrinsically bounded within a specific range, whereas GMMs have unbounded support over . This mismatch in support may lead to suboptimal fitting. By normalizing the KL divergence values, we map them to , which precisely matches the natural bounded support of the Beta distribution. Second, in OSGDA, the presence of unknown categories inevitably introduces open-set noise. Under such noisy conditions, the EM algorithm used to optimize GMMs exhibits reduced stability and is prone to becoming trapped in poor local optima, preventing reliable fitting. In contrast, parameter estimation for the Beta distribution maintains high stability and robustness to such noise.
In summary, We adopt the Beta distribution for distribution modeling, aiming to reliably and consistently identify high-quality unknown-class nodes. We hope this explanation addresses the reviewer's concerns.
Q6: About the Edge Down-weighting Strategy.
R6: Thank you for the comment! Incorrectly identifying known-class nodes as unknown may reduce the contribution of useful neighborhood information during subsequent message passing. Therefore we incorporate nonzero edge down-weighting to provide tolerance for such misidentification. Specifically, in Eq. (14), we multiply the weights of edges connected to predicted unknown-class nodes by a small fixed value, while leaving the remaining edge weights unchanged. Thus, even when a known-class node is misidentified, its connections can still participate in message passing with reduced weights. This preserves information propagation while suppressing the influence of predicted unknown nodes, thereby partially mitigating the issue you raised. Furthermore, we also conduct experiments with different perturbation coefficient for edges connected to predicted unknown nodes, with the results (ACC/HS) presented below.
| A2D | A2C | C2D | |
|---|---|---|---|
| {0,1} | 71.26/70.34 | 73.44/72.46 | 71.58/71.02 |
| {0.1,1} | 74.70/74.10 | 75.68/74.77 | 75.68/73.34 |
| {0.2,1} | 74.69/74.14 | 75.49/74.95 | 75.73/73.22 |
| {0.5,1} | 72.54/71.83 | 74.53/73.26 | 74.21/72.01 |
| {0.8,1} | 70.68/70.01 | 73.24/72.89 | 71.30/70.88 |
The results show that small nonzero perturbation coefficients yield favorable performance, whereas setting the coefficient to 0 leads to a performance decline. This indicates that the issue the reviewer raised does arise. We therefore retain a small nonzero coefficient to reduce the impact of incorrectly identifying known-class nodes as unknown during training.
We thank the reviewer for the constructive feedback, which helpes us improve the quality of our work. All revisions mentioned above have been incorporated into the revised manuscript and marked in red, please refer to it for details. We hope that our responses satisfactorily address the reviewer’s concerns.
We sincerely thank you for your time and effort in reviewing our manuscript and for providing valuable feedback. We carefully address your comments and concerns in the following responses.
R1: Thank you for your valuable comment! To further evaluate the performance of our DEAR and the OSGDA baselines (SDA[1], UAGA[2], ETA[3], and GraphRTA[4]), we conduct additional experiments on six transfer tasks, treating classes 0 and 1 as unknown classes, and each experiment is repeated five times, and we report the mean and standard deviation of known-class classification accuracy (OS∗) and AUC. The results are presented below (with the best score highlighted in bold):
As shown above, existing OSGDA methods generally achieve competitive performance in known-class classification, while our DEAR maintains comparable OS∗ performance and demonstrates particularly strong results in AUC, achieving the best performance on most transfer tasks. This indicates that DEAR is more effective at distinguishing known-class nodes from unknown-class nodes across different decision thresholds. We attribute this improvement to the virtual-class guided boundary learning, which encourages more discriminative representations of known classes, together with the KL divergence based unknown-class identification mechanism, which facilitates the separation of known and unknown target nodes. These findings further demonstrate the effectiveness of our method in OSGDA tasks.
We hope these additional experiments adequately address your concerns.
[1] Open-set graph domain adaptation via separate domain alignment, 2024 AAAI.
[2] Open-Set Cross-Network Node Classification via Unknown-Excluded Adversarial Graph Domain Alignment, 2025 AAAI.
[3] ETA: Dual Evidence-Aware Uncertainty Learning for Open-Set Graph Domain Adaptation, 2025.
[4] Towards Unsupervised Open-Set Graph Domain Adaptation via Dual Reprogramming, 2025 NeurIPS.
R2: Thank you for your insightful question! The improved source-domain decision boundaries facilitate unknown-class recognition by enhancing the discriminability of known-class representations. As analyzed in Theorem 1, virtual-class learning introduces additional feature-alignment supervision that facilitates classification margin optimization, encouraging more discriminative known-class representations and source-class prediction prototypes. These prototypes serve as references for target-domain recognition, where known-class nodes tend to exhibit prediction distributions similar to their corresponding source prototypes, while unknown-class nodes tend to show larger KL divergences, facilitating their identification.
This mechanism is particularly effective when known classes retain transferable semantic information across domains and the prediction distributions of known and unknown nodes exhibit sufficient separation. Furthermore, the margin visualizations in Figure 7 and the ablation results in Figure 3 further provide empirical support for the effectiveness of the virtual-class guided boundary learning.
We have further clarified this mechanism in Section 2.3 of the revised manuscript, highlighting the role of source-domain representations in target-domain unknown-class identification. We hope this explanation adequately addresses your concerns.