Research map / Image generation / Autoregressive and latent generation
An Iterative Polishing Framework Based on Quality Aware Masked Language Model for Chinese Poetry Generation
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 |
|---|---|---|---|
| Polish AS2S poems with QA-MLM | AS2S · Chinese quatrain test | Meaningfulness: 2.35 -> 2.59 | Helped |
| Polish CVAE-D poems with QA-MLM | CVAE-D · Chinese quatrain test | Meaningfulness: 2.55 -> 2.74 | Helped |
| Polish B&T poems with QA-MLM | B&T · Chinese quatrain test | Meaningfulness: 3.06 -> 3.24 | Helped |
| Polish AS2S poems with QA-MLM | AS2S · Chinese quatrain test | BLEU-1: 0.032 -> 0.042 | Helped |
| Polish CVAE-D poems with QA-MLM | CVAE-D · Chinese quatrain test | BLEU-1: 0.039 -> 0.039 | No clear effect |
| Polish B&T poems with QA-MLM | B&T · Chinese quatrain test | BLEU-1: 0.050 -> 0.049 | Hurt |
| Polish B&T poems with QA-MLM | B&T · Chinese quatrain test | Tone accuracy: 0.976 -> 0.962 | Hurt |
| Polish B&T poems with QA-MLM | B&T · Chinese quatrain test | Rhythm accuracy: 0.956 -> 0.841 | Hurt |
| Polish poems with QA-MLM | AS2S, CVAE-D, and B&T · Chinese quatrain test | Similarity | Helped |
| Use global left-and-right context to replace low-quality characters rather than sequential decoding alone | Encoder-decoder poem drafts · Chinese poetry generation | Human-rated poem quality | Helped |
Closest papers by meaning
- PoetryDiffusion: Towards Joint Semantic and Metrical Manipulation in Poetry Generation (AAAI 2024 · similarity 0.86)
- Turning Up the Heat: Min-p Sampling for Creative and Coherent LLM Outputs (ICLR 2025 · similarity 0.83)
- A Multi-Style Chinese Characters Writing Intelligent Tool Based on Small-scale Training Data (AAAI 2025 · similarity 0.83)
- Quality-Diversity through AI Feedback (ICLR 2024 · similarity 0.82)
- Top-H Decoding: Adapting the Creativity and Coherence with Bounded Entropy in Text Generation (NeurIPS 2025 · similarity 0.82)
- WenMind: A Comprehensive Benchmark for Evaluating Large Language Models in Chinese Classical Literature and Language Arts (NeurIPS 2024 · 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.