Research map / Label-efficient and robust learning / Recognition and long-tail
Dual-Clustering Maximum Entropy with Application to Classification and Word Embedding
Xiaolong Wang, Jingjing Wang, Chengxiang Zhai
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 |
|---|---|---|---|
| use online/offline tuning with Q=10 instead of Q=0 (ablation) | DCME · NYT word embedding | log-likelihood (overall): −8.663 -> −8.654 | Improved |
| use online/offline tuning with Q=10 instead of Q=0 (ablation) | DCME · NYT word embedding | time per epoch (minutes): 25.59 -> 33.49 | Hurt |
| use online/offline tuning with Q=10 instead of Q=0 (ablation) | DCME · ACM Digital Library classification | test accuracy | Improved |
| replace NCE with DCME-Q10 | word embedding · Semantic-Syntactic Word Relationship | log-likelihood (overall): −8.783 -> −8.654 | Improved |
| replace NS with DCME-Q10 | word embedding · Semantic-Syntactic Word Relationship | log-likelihood (overall): −8.699 -> −8.654 | Improved |
| replace gradient descent with DCME-Q10 | ME classification · ACM Digital Library classification | time per epoch (seconds): 165.91 -> 9.50 | Improved |
| replace gradient descent with DCME for accuracy | ME classification · ACM Digital Library classification | test accuracy | Hurt |
| replace sampling (NCE/NS) with DCME | ME classification · ACM Digital Library classification | test accuracy | Improved |
| replace DCME-Q10 with NCE for speed | ME classification · ACM Digital Library classification | time per epoch (seconds): 9.50 -> 6.21 | Improved |
Closest papers by meaning
- SEC: More Accurate Clustering Algorithm via Structural Entropy (AAAI 2024 · similarity 0.83)
- Learning to Embed Distributions via Maximum Kernel Entropy (NeurIPS 2024 · similarity 0.83)
- Decoupled Entropy Minimization (NeurIPS 2025 · similarity 0.83)
- The Entropy Enigma: Success and Failure of Entropy Minimization (ICML 2024 · similarity 0.83)
- Repurposing Language Models into Embedding Models: Finding the Compute-Optimal Recipe (NeurIPS 2024 · similarity 0.83)
- Cut Your Losses in Large-Vocabulary Language Models (ICLR 2025 · similarity 0.82)
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.
