Research map / Label-efficient and robust learning / Recognition and long-tail
CLARE: A Joint Approach to Label Classification and Tag Recommendation
Yilin Wang, Suhang Wang, Jiliang Tang, Guojun Qi, Huan Liu, Baoxin Li
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 |
|---|---|---|---|
| jointly model classification and tag recommendation with label-tag bipartite group constraints (CLARE) vs best baseline | best baseline (LS/FT) · NUS-Flickr | accuracy: 35.67% -> 40.32% | Improved |
| CLARE vs best baseline | best baseline (LS) · Shoe-Zappo | accuracy: 86.03% -> 89.39% | Improved |
| CLARE vs best baseline | best baseline (LS) · USAA-YouTube | accuracy: 39.39% -> 53.07% | Improved |
| CLARE vs best baseline on tag recommendation | best baseline (FT) · USAA-YouTube | AP@10: 51.78% -> 62.95% | Improved |
| CLARE vs LS on tag recommendation AP@3 | LS · NUS-Flickr | AP@3: 22.69% -> 21.22% | Hurt |
| model label-tag bipartite graph vs GLasso that ignores it | GLasso · Shoe-Zappo | accuracy: 76.31% -> 89.39% | Improved |
| increase alpha (label-tag relation component weight) from 10^-6 to 0.01 (ablation) | CLARE · Shoe-Zappo | performance: almost 8% | Improved |
| increase alpha from 10 to 100 (ablation) | CLARE · Shoe-Zappo | performance | Hurt |
| use alpha in [0.3, 1] (ablation) | CLARE · Shoe-Zappo | performance | No clear effect |
| use very small or very large beta (ablation) | CLARE · Shoe-Zappo | performance | Hurt |
Later papers that compared with this method
1 later accepted paper reports CLARE as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.
- SLRL: Semi-Supervised Local Community Detection Based on Reinforcement Learning (AAAI 2025 · better on 1 of 1 shared result)
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
- An Integral Tag Recommendation Model for Textual Content (AAAI 2019 · similarity 0.86)
- Capturing Semantic Correlation for Item Recommendation in Tagging Systems (AAAI 2016 · similarity 0.85)
- Improving Recommendation of Tail Tags for Questions in Community Question Answering (AAAI 2016 · similarity 0.84)
- An Interactive Multi-Label Consensus Labeling Model for Multiple Labeler Judgments (AAAI 2018 · similarity 0.83)
- How Does Message Passing Improve Collaborative Filtering? (NeurIPS 2024 · similarity 0.82)
- Multi-Label Node Classification with Label Influence Propagation (ICLR 2025 · 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.
