Research map / Label-efficient and robust learning / Recognition and long-tail
Distributional Semantics Meets Multi-Label Learning
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| replace SLEEC's iterative SVP embedding learning with single SPPMI factorization | EXMLDS1 · Bibtex | training time (s): 313 -> 23 | Helped |
| replace SLEEC embedding learning with SPPMI factorization | EXMLDS1 · Delicious-200K | training time (s): 10000 -> 1937 | Helped |
| use SPPMI matrix factorization instead of word2vec neural network for the same objective (ablation) | EXMLDS1 vs EXMLDS2 · Delicious-200K | training time (s): 13000 -> 1937 | Helped |
| SPPMI factorization vs word2vec network for embeddings (ablation) | EXMLDS1 vs EXMLDS2 · small benchmarks | P@k | No clear effect |
| use unweighted SVD on PPMI rather than SLEEC's embedding learning | EXMLDS1 · Bibtex, Eurlex, Delicious-200K | P@1: 65.29 -> 63.38 (Bibtex) | Hurt |
| add label-correlation objective and label embeddings (EXMLDS3) when 80% labels hidden | EXMLDS1-style embedding · Bibtex 80% hidden | P@1: 30.5 -> 48.51 (vs SLEEC) | Helped |
| add label-correlation objective (EXMLDS3) vs LEML-IMC using same information | missing-label setting · Eurlex 80% hidden | P@1: 39.24 -> 60.28 | Helped |
| jointly learn embeddings and regressor via gradient descent (EXMLDS4) | EXMLDS1 · large datasets | P@1 | Helped |
| EXMLDS4 joint learning vs ANNEXML | EXMLDS4 · WikiLSHTC-325K | P@1: 63.36 -> 62.15 | Hurt |
Closest papers by meaning
- Modeling Label Space Interactions in Multi-label Classification using Box Embeddings (ICLR 2022 · similarity 0.88)
- Exploiting Negative Multi-Cluster Structure in Class-Wise Embeddings for Weakly Supervised Multi-Label Learning (NeurIPS 2026 · similarity 0.88)
- Compositional Generalization for Multi-Label Text Classification: A Data-Augmentation Approach (AAAI 2024 · similarity 0.86)
- Predicting Label Distribution from Ternary Labels (NeurIPS 2024 · similarity 0.86)
- Adaptive Integration of Partial Label Learning and Negative Learning for Enhanced Noisy Label Learning (AAAI 2024 · similarity 0.86)
- Harnessing the Power of Beta Scoring in Deep Active Learning for Multi-Label Text Classification (AAAI 2024 · similarity 0.85)
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.