Research map / Image generation / Autoregressive and latent generation
Generating Chinese Ci with Designated Metrical Structure
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 |
|---|---|---|---|
| add explicit metrical (Cipai) condition vs MRCG- without it (ablation) | MRCG · Guided generation | length accuracy: 41.96 -> 99.37 | Helped |
| add explicit metrical condition (ablation) | MRCG · Guided generation | tone accuracy: 66.21 -> 93.71 | Helped |
| add explicit metrical condition (ablation) | MRCG · Guided generation | rhyme accuracy: 39.77 -> 98.28 | Helped |
| add explicit metrical condition (ablation) | MRCG · Blind generation | length accuracy: 35.49 -> 99.21 | Helped |
| add explicit metrical condition (ablation) | MRCG · Blind generation | tone accuracy: 59.43 -> 92.03 | Helped |
| add explicit metrical condition (ablation) | MRCG · Blind generation | rhyme accuracy: 33.34 -> 96.87 | Helped |
| use MRCG instead of Seq2seq baseline | MRCG · Guided generation | length accuracy: 27.1 -> 99.37 | Helped |
| use MRCG instead of attention baseline | MRCG · Guided generation | rhyme accuracy: 26.7 -> 98.28 | Helped |
| add metrical condition | MRCG · Ci semantics | RST Pearson correlation: 0.541 vs 0.529; Seq2Seq 0.358 | Helped |
| add metrical condition (MRCG- to MRCG) (ablation) | MRCG · Ci semantics | AST BLEU: 0.235 -> 0.229 | No clear effect |
| add metrical condition (ablation) | MRCG · Human evaluation | fluency: 2.38 -> 2.59 | Helped |
| add metrical condition (ablation) | MRCG · Human evaluation | aesthetics: 2.34 -> 2.54 | Helped |
Closest papers by meaning
- PoetryDiffusion: Towards Joint Semantic and Metrical Manipulation in Poetry Generation (AAAI 2024 · similarity 0.86)
- WenMind: A Comprehensive Benchmark for Evaluating Large Language Models in Chinese Classical Literature and Language Arts (NeurIPS 2024 · similarity 0.81)
- Whole-Song Hierarchical Generation of Symbolic Music Using Cascaded Diffusion Models (ICLR 2024 · similarity 0.80)
- CSL-L2M: Controllable Song-Level Lyric-to-Melody Generation Based on Conditional Transformer with Fine-Grained Lyric and Musical Controls (AAAI 2025 · similarity 0.80)
- S²MILE: Semantic-and-Structure-Aware Music-Driven Lyric Generation (AAAI 2025 · similarity 0.80)
- SongGLM: Lyric-to-Melody Generation with 2D Alignment Encoding and Multi-Task Pre-Training (AAAI 2025 · similarity 0.79)
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.