- Rating
- 2
- Confidence
- 2
- Informativeness
- 2
- Presentation
- 2
- Soundness
- 2
Read review
Paper summary
This paper proposes GBASC, a deep multi-view clustering framework that addresses structural errors through adaptive granular-ball splitting and connection. GBAS evaluates candidate splits using multiple criteria, including compactness, neighborhood structure, cross-view agreement, stability, and spectral guidance. GBAC assesses the reliability of inter-ball connections, attenuates high-risk relations, and selects global cluster assignments from multiple candidates using additional quality and trust checks. The resulting clustering structure further guides representation learning. Extensive experiments demonstrate strong clustering performance across multiple benchmarks, supported by ablation studies and structural diagnostics.
Critical strengths
-
Strong and consistent empirical performance. GBASC achieves the best reported ACC, NMI, and ARI across all six main benchmarks against nine baselines, with further competitive results in the extended evaluations.
-
Comprehensive and targeted experimental validation. The paper provides factorial ablations and targeted diagnostics of splitting quality, risk attenuation, and partition selection.
Critical weaknesses
1. Methodological Novelty and Design Justification.
While GBASC introduces new mechanisms, its overall refinement-and-connection framework builds on existing granular-ball clustering methods (e.g., GBCT and MGBCC). It involves numerous heuristic choices, such as fixed splitting-evidence weights and additional spectral-guided decision rules in GBAS, and weighted candidate scoring and trust checks in GBAC. Although intuitively motivated, their specific formulations and individual contributions remain insufficiently justified, while the theoretical results mainly establish general or conditional properties rather than directly supporting these design choices.
2. Unclear Practical Viability.
GBASC involves numerous fixed hyperparameters, including splitting weights, complexity penalties, and risk attenuation coefficients, without systematic sensitivity analysis of the final clustering performance, leaving its tuning requirements unclear. Moreover, repeated graph construction, spectral decomposition, and candidate partition evaluation may incur substantial computational costs, yet neither an overall complexity analysis nor runtime comparisons with baselines are provided.
Questions for authors
-
Could the authors further clarify the contributions of the key mechanisms within GBAS and GBAC? For example, by isolating risk attenuation from five-candidate selection (and clarifying their settings in Table 3), or evaluating the contribution of terminal spectral-guided splitting. Targeted comparisons on representative datasets would help establish which mechanisms drive the performance gains.
-
The proposed method involves numerous manually specified hyperparameters, such as the GBAS splitting weights and the risk attenuation coefficient β. Could the authors provide brief sensitivity analysis or other evidence showing that the clustering performance is robust to these hyperparameter choices?
-
Given the repeated graph construction, spectral decomposition, and multiple candidate partition evaluations in GBASC, could the authors provide an overall computational complexity analysis and end-to-end runtime comparisons with representative baselines on ALOI (over 10,000 samples)?
5 comments
Answer to Question 2
We thank the reviewer for this valuable suggestion. We conducted sensitivity experiments on ORL and DHA to evaluate the robustness of GBASC to the risk attenuation coefficient β and the GBAS candidate-evidence weights (ω_m, ω_f, ω_a). Each configuration was evaluated using 10 random seeds (42–51), with all other settings fixed. Results are reported as mean ± std (%).
1. Sensitivity to β.
Following Eqs. (7) and (11), β controls risk attenuation in graph construction and reliability-weighted structural feedback. We varied β ∈ {0, 0.10, 0.25, 0.50, 1.00}, where β = 0 serves as a no-attenuation control, while fixing the GBAS weights at their default values.
Table 1. Sensitivity to risk attenuation coefficient β.
| Dataset | β | ACC ↑ | NMI ↑ | ARI ↑ | Purity ↑ |
|---|---|---|---|---|---|
| ORL | 0.00 | 83.55±1.23 | 91.20±0.55 | 76.16±1.57 | 84.53±1.11 |
| ORL | 0.10 | 83.40±1.23 | 90.98±0.55 | 75.45±1.50 | 84.10±0.96 |
| ORL | 0.25* | 83.22±1.13 | 91.02±0.58 | 75.64±1.41 | 84.35±1.01 |
| ORL | 0.50 | 83.62±0.99 | 91.08±0.45 | 75.87±1.09 | 84.55±0.84 |
| ORL | 1.00 | 83.92±1.51 | 91.40±0.65 | 76.64±1.77 | 84.95±1.52 |
| DHA | 0.00 | 86.44±0.15 | 88.91±0.15 | 79.36±0.24 | 87.23±0.14 |
| DHA | 0.10 | 86.46±0.11 | 88.96±0.13 | 79.46±0.21 | 87.29±0.11 |
| DHA | 0.25* | 86.44±0.18 | 88.92±0.20 | 79.37±0.32 | 87.23±0.20 |
| DHA | 0.50 | 86.34±0.14 | 88.77±0.15 | 79.15±0.23 | 87.10±0.14 |
| DHA | 1.00 | 86.21±0.26 | 88.58±0.22 | 78.83±0.39 | 86.98±0.18 |
*Default setting.
Across the tested β values, ACC varies by at most 0.70 percentage points on ORL and 0.25 on DHA, with similarly limited variations in NMI, ARI, and Purity. These results indicate that GBASC maintains stable clustering performance without requiring precise tuning of β within the investigated range.
2. Sensitivity to GBAS candidate-evidence weights.
As defined in Eq. (3), (ω_m, ω_f, ω_a) weight compactness improvement, neighborhood fracture, and cross-view agreement when ranking candidate splits. We evaluated four configurations while fixing β = 0.25.
Table 2. Sensitivity to GBAS candidate-evidence weights.
| Dataset | (ω_m, ω_f, ω_a) | ACC ↑ | NMI ↑ | ARI ↑ | Purity ↑ |
|---|---|---|---|---|---|
| ORL | (0.45,0.30,0.25)* | 83.22±1.13 | 91.02±0.58 | 75.64±1.41 | 84.35±1.01 |
| ORL | (1/3,1/3,1/3) | 82.98±1.59 | 91.00±0.63 | 75.27±1.64 | 84.10±1.33 |
| ORL | (0.35,0.40,0.25) | 84.03±1.19 | 91.34±0.49 | 76.40±1.29 | 84.65±1.06 |
| ORL | (0.55,0.25,0.20) | 84.05±1.00 | 91.23±0.46 | 76.25±1.28 | 84.90±0.93 |
| DHA | (0.45,0.30,0.25)* | 86.44±0.18 | 88.92±0.20 | 79.37±0.32 | 87.23±0.20 |
| DHA | (1/3,1/3,1/3) | 86.29±0.16 | 88.69±0.14 | 78.99±0.22 | 87.06±0.11 |
| DHA | (0.35,0.40,0.25) | 86.48±0.14 | 88.95±0.14 | 79.43±0.22 | 87.27±0.11 |
| DHA | (0.55,0.25,0.20) | 86.31±0.15 | 88.82±0.12 | 79.14±0.20 | 87.10±0.10 |
*Default setting.
Changing the evidence weights produces maximum ACC variations of only 1.08 percentage points on ORL and 0.19 on DHA. The remaining metrics also exhibit limited fluctuations, suggesting that candidate ranking is not critically dependent on a particular weight allocation.
3. Complementary ablation evidence.
Due to space limitations, we summarize the component ablations without reproducing the existing ablation table. As reported in Table 3 of the manuscript, on DHA, adaptive splitting improves ACC from 78.20% to 85.70% under P-Connection. On ALOI, replacing P-Connection with Risk Connection under adaptive splitting increases ACC from 90.60% to 96.85% and ARI from 87.25% to 93.85%. These results demonstrate the complementary contributions of GBAS and GBAC, while the new sensitivity experiments show that performance remains relatively stable under the tested parameter variations.
We will incorporate these sensitivity results into Section 4.3 and clarify the parameter definitions and default settings in Sections 3.2–3.3, including the shared use of β in structural feedback.
We sincerely thank the reviewer for this constructive suggestion.
Answer to Question 3
We sincerely thank the reviewer for this valuable suggestion. We agree that computational complexity and end-to-end runtime are important for evaluating GBASC's scalability. We provide the following analysis and additional experimental results, which will be incorporated into the revised manuscript.
1. Computational Complexity
Let , , , and denote the numbers of samples, views, total feature dimensions, and resulting atomic granular balls, respectively.
The main computational costs arise from: (1) multi-view graph construction and Ward initialization, requiring ; (2) GBAS local refinement, requiring with precomputed graphs, where the maximum ball size and split depth are bounded; and (3) GBAC quotient-graph spectral decomposition, requiring . Importantly, GBAC evaluates only five candidate partitions with bounded optimization iterations, avoiding combinatorial partition enumeration.
Including two structural evaluations and training epochs, the overall leading complexity is
where denotes per-sample training cost, denotes sample-level sparse spectral solving cost, and includes local refinement and implementation overhead.
Under fixed configurations and controlled auxiliary costs, the dominant complexity simplifies to . This is not an unconditional worst-case bound. GBASC's efficiency primarily comes from performing global assignments on atoms instead of samples, while restricting local refinement and candidate evaluation.
2. End-to-End Runtime on ALOI
We additionally report end-to-end runtimes on ALOI (10,800 samples, 4 views, 100 clusters).
| Method | Runtime (s) ↓ | Speedup |
|---|---|---|
| DCMVC | 2411.41 ± 6.20 | 3.66× |
| CorreGen | 2969.67 ± 477.06 | 4.51× |
| RAV | 4257.64 ± 1025.91 | 6.47× |
| RAC-DMVC | 2277.13 ± 401.56 | 3.46× |
| GBASC (Ours) | 658.44 ± 1.22 | 1.00× |
GBASC completes clustering in 658.44 seconds (10.97 minutes), achieving 3.46–6.47× speedups over these representative baselines. Meanwhile, it achieves 96.85% ACC, 97.04% NMI, and 93.85% ARI, demonstrating a favorable balance between clustering accuracy and computational efficiency.
3. Revision Commitment
We appreciate the reviewer's constructive feedback. In the revised manuscript, we will add (i) a detailed module-wise complexity analysis with explicit assumptions, (ii) end-to-end runtime comparisons on ALOI, and (iii) hardware specifications and timing protocols for reproducibility. We believe these additions will strengthen the paper's evaluation of computational efficiency and scalability.
Answer to Question 1
We thank the reviewer for this valuable suggestion. We agree that our original module-level ablation in Table 3 does not fully disentangle the contributions of individual decision mechanisms within GBAS and GBAC.
1. Distinct roles of the decision mechanisms.
In GBAS, candidate evidence is used to rank proposed splits, whereas the subsequent acceptance decision controls whether a split should be executed. The cost criterion discourages excessive fragmentation, while the structure criterion permits boundaries supported by global structural evidence. The terminal spectral-guided refinement further addresses boundaries that may not be represented in the local candidate pool.
In GBAC, risk attenuation calibrates inter-atom affinities before obtaining the base partition. Five candidate slots provide alternative global assignments, while trust constraints regulate whether the best-scoring nonbase candidate can safely replace the base partition. If the acceptance conditions are not satisfied, the method retains the base partition.
2. Fine-grained component ablations.
To clarify these contributions, we conduct targeted comparisons on ORL, DHA, and ALOI under consistent evaluation protocols.
For GBAS, we distinguish cost-only acceptance (A1), structure-only acceptance (A2), dual-gate acceptance without terminal guidance (A3), and full GBAS with terminal spectral guidance (A4).
The results indicate that A3 consistently outperforms either single-criterion variant across the three datasets. Adding terminal spectral guidance further improves ALOI ACC from 95.50% to 96.85% and reduces pairwise error from 0.14% to 0.12%. These results support the complementary roles of local acceptance and terminal refinement.
For GBAC, we separately examine risk attenuation, five-candidate selection, and trust-constrained acceptance.
On ALOI, enabling risk attenuation while retaining base-only assignment (B1 vs. B3) increases ACC from 90.20% to 95.50%, with pairwise error decreasing from 0.29% to 0.16%. In contrast, enabling candidate selection without risk attenuation (B1 vs. B2) improves ACC by only 0.60 percentage points.
With risk attenuation enabled, unconstrained candidate selection (B4) achieves 95.10% ACC and 0.18% pairwise error, whereas trust-constrained selection (B5) reaches 96.85% ACC and 0.12% pairwise error. This comparison suggests that trust constraints help prevent harmful candidate replacements while retaining beneficial ones.
Together, these comparisons indicate that the components play distinct and complementary roles, rather than merely adding heuristic complexity.
3. Clarification of experimental settings.
We will explicitly document the activation of risk attenuation, candidate selection, and trust constraints in Table 3, including which settings remain fixed across the P-Connection and Risk Connection variants.
We will also revise Sections 3.2–3.3 to distinguish candidate ranking, split acceptance, terminal refinement, partition selection, and fallback. The fine-grained comparisons will be incorporated into Section 4.3 and the supplementary material.
We appreciate the reviewer's suggestion, which helps us present a more precise explanation of the contributions of the individual mechanisms.