AI Paper Atlas

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

ICML 2026 · Multi-agent systems and games › Human-AI and goal-conditioned · approach: World-model-based agents

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 changedWhereMeasuredEffect
add Auto-K agent to backbone (ablation)GROK · Caltech7ACC: 0.335 -> 0.622Improved
add Auto-K agent to backbone (without structure guidance) (ablation)GROK · HdigitACC: 0.982 -> 0.900Hurt
add Auto-K agent to backbone (without structure guidance) (ablation)GROK · CCVACC: 0.343 -> 0.284Hurt
add geometric structure guidance on top of Auto-K agent (ablation)GROK · CCVACC: 0.284 -> 0.369Improved
add adaptive reward normalization versus none (ablation)GROK · DHApredicted K: 39 -> 22 (K_GT 23)Improved
add adaptive normalization versus unified normalization (ablation)GROK · DHAACC: 0.261 -> 0.816Improved
use multi-agent divide-and-conquer instead of single agent (ablation)GROK · DHApredicted K: 4 -> 22Improved
use temperature-based sampling instead of top-k random baseline (ablation)GROK · Hdigitconvergence timeImproved
increase group sampling number NS (ablation)GROK · Hdigitconvergence of KImproved
use GROK instead of DeepDPMGROK · CIFAR10ACC: 0.651 -> 0.993Improved
use GROK agent instead of known K on DHA / YouTubeFaceGROK · DHAACC: 0.825 -> 0.816 (AdaM vs GROK)Hurt
add cluster decision agent as plug-and-play to MFLVC, GCFAggMVC, SEMcluster decision agent · Caltech7ACC: up to 13.6% ACC improvementImproved

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.

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

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy