Research map / Multi-agent systems and games / Human-AI and goal-conditioned
GRPO-based Cluster Decision Agent for Unknown-$\boldsymbol{K}$ Multi-view Clustering
Xuqian Xue, Jun Zhang, Qi Cai, Zhizhong Huang, Hongming Shan, Junping Zhang
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add Auto-K agent to backbone (ablation) | GROK · Caltech7 | ACC: 0.335 -> 0.622 | Improved |
| add Auto-K agent to backbone (without structure guidance) (ablation) | GROK · Hdigit | ACC: 0.982 -> 0.900 | Hurt |
| add Auto-K agent to backbone (without structure guidance) (ablation) | GROK · CCV | ACC: 0.343 -> 0.284 | Hurt |
| add geometric structure guidance on top of Auto-K agent (ablation) | GROK · CCV | ACC: 0.284 -> 0.369 | Improved |
| add adaptive reward normalization versus none (ablation) | GROK · DHA | predicted K: 39 -> 22 (K_GT 23) | Improved |
| add adaptive normalization versus unified normalization (ablation) | GROK · DHA | ACC: 0.261 -> 0.816 | Improved |
| use multi-agent divide-and-conquer instead of single agent (ablation) | GROK · DHA | predicted K: 4 -> 22 | Improved |
| use temperature-based sampling instead of top-k random baseline (ablation) | GROK · Hdigit | convergence time | Improved |
| increase group sampling number NS (ablation) | GROK · Hdigit | convergence of K | Improved |
| use GROK instead of DeepDPM | GROK · CIFAR10 | ACC: 0.651 -> 0.993 | Improved |
| use GROK agent instead of known K on DHA / YouTubeFace | GROK · DHA | ACC: 0.825 -> 0.816 (AdaM vs GROK) | Hurt |
| add cluster decision agent as plug-and-play to MFLVC, GCFAggMVC, SEM | cluster decision agent · Caltech7 | ACC: up to 13.6% ACC improvement | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- AdaM — ADAM: An Embodied Causal Agent in Open-World Environments (ICLR 2025) · better on 2 of 3 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Multi-View Clustering by Inter-cluster Connectivity Guided Reward (ICML 2024 · similarity 0.89)
- EASEMVC:Efficient Dual Selection Mechanism for Deep Multi-View Clustering (CVPR 2025 · similarity 0.86)
- Medusa: A Multi-Scale High-order Contrastive Dual-Diffusion Approach for Multi-View Clustering (CVPR 2025 · similarity 0.85)
- An Effective Augmented Lagrangian Method for Fine-Grained Multi-View Optimization (AAAI 2024 · similarity 0.84)
- Max-Mahalanobis Anchors Guidance for Multi-View Clustering (AAAI 2025 · similarity 0.84)
- Evaluate then Cooperate: Shapley-based View Cooperation Enhancement for Multi-view Clustering (NeurIPS 2024 · similarity 0.84)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
