Research map / Label-efficient and robust learning / Recognition and long-tail
Discovering New Intents via Constrained Deep Adaptive Clustering with Cluster Refinement
Ting-En Lin, Hua Xu, Hanlei 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 pairwise constraints from limited labeled data (ablation) | CDAC+ · SNIPS | ACC: 91.41 -> 93.63 | Improved |
| add pairwise constraints from limited labeled data (ablation) | CDAC+ · DBPedia | ACC: 82.88 -> 91.66 | Improved |
| add pairwise constraints from limited labeled data (ablation) | CDAC+ · StackOverflow | ACC: 23.69 -> 73.48 | Improved |
| add cluster refinement with KLD loss (ablation) | CDAC · SNIPS | ACC: 74.93 -> 93.63 | Improved |
| add cluster refinement with KLD loss (ablation) | CDAC · DBPedia | ACC: 69.01 -> 91.66 | Improved |
| add cluster refinement with KLD loss (ablation) | CDAC · StackOverflow | ACC: 23.97 -> 73.48 | Improved |
| use cluster refinement instead of K-means on learned representations (ablation) | CDAC+ · SNIPS | ACC: 93.03 -> 93.63 | Improved |
| use cluster refinement instead of K-means on learned representations (ablation) | CDAC+ · DBPedia | ACC: 89.77 -> 91.66 | Improved |
| use cluster refinement instead of K-means on learned representations (ablation) | CDAC+ · StackOverflow | ACC: 71.49 -> 73.48 | Improved |
| increase labeled-data ratio | CDAC+ · StackOverflow | NMI | Improved |
| increase unknown-class ratio | BERT-Semi · datasets with more unknown classes | clustering performance | Hurt |
Closest papers by meaning
- End-to-end Learnable Clustering for Intent Learning in Recommendation (NeurIPS 2024 · similarity 0.85)
- Interactive Deep Clustering via Value Mining (NeurIPS 2024 · similarity 0.83)
- Breaking the Reclustering Barrier in Centroid-based Deep Clustering (ICLR 2025 · similarity 0.82)
- Multi-Granularity Open Intent Classification via Adaptive Granular-Ball Decision Boundary (AAAI 2025 · similarity 0.82)
- Personalized Clustering via Targeted Representation Learning (AAAI 2025 · similarity 0.82)
- MIntRec2.0: A Large-scale Benchmark Dataset for Multimodal Intent Recognition and Out-of-scope Detection in Conversations (ICLR 2024 · similarity 0.81)
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.
